Pierre Collet – A diachronic epistemology of Complex Systems and their engineering

Pierre Collet

The Faculty of Engineering, Universidad Alberto Hurtado, kindly invites you to the talk of our honored guest, Professor Pierre Collet (Strasbourg University, France) “A diachronic epistemology of Complex Systems and their engineering”.

Pierre Collet (57) is a distinguished Professor in Computer Science at Strasbourg University (France). Pierre Collet has published around 200 refereed research papers. He specializes on Complex Systems, Artificial Intelligence, massively parallel artificial evolution, emergent e-education, epistemology and philosophy. His is currently researching whether the developing artificial entities require a new vision of meta-ethics.

Sorry so good morning everyone um on behalf of the faculty of Engineering University Alber we are pleased to welcome and introduce we are pleased to welcome and introduce our honored guest professor Pier kle um he is um distinguished Professor uh in computer science at Strasburg University um just to mention some

Highlights of his great career uh from 2003 first thing is oh this is a bit loud I was born in chid that is the first highlight of my career but I think after a month we but I left after one month so um I’m sorry but I cannot do

This presentation but that’s a highest thing in my career explain a lot of things that you will present the complex maybe thing so from uh 2003 to 2008 he was the president of the artificial of the French Association for artificial Evolution from uh 2011 to

2015 H he was the head of the computer science department at Strasburg University and during those years he created the complex systems and translational bioinformatics research group and he co-created the complex systems digital campus UNESCO uniwin which is a network of uh University that are developing research and education on complex

Systems and professor kle kindly invited us to join that Network so we really thank you for that for that great opportunity in 201 16 he um co-founded the uh new French aerb University so um Pier’s interests include complex systems artificial intelligence and ethics uh artificial Evolution Innovation for Education among

Others and and when uh we were planning this event we realized the particular importance of your visit and and this talk for us because we are a really young uh engineering faculty of engineering uh in in this in this University um which has a a vast uh tradition in social sciences and

Humanities and I think that your approach roach really uh Bridges the gap between engineering and philosophy so we have a lot to learn from you from your experience and how to build this bridge thank you very much Pier for being here okay uh thank you very much and

Thank you for this really nice introduction um we had the opportunity to talk together uh because for the next conference um that will come on artificial intelligence and ethics uh we translated it to Spanish for the students and by talking with christobal I understood that you had a big

Understanding what complex systems are and so what I’m going to present here is um the title I gave uh was a dionic epistemology of complex system dionic means through the times and um I will show how complex systems uh were uh conceived and created along the ages

And it will start with uh Greek philosophy okay all right so uh let’s say it is quarter past 9 so quarter 10 so let’s yeah okay all right thank you um okay so uh yes as uh christobal said uh I’m the co-coordinator of the complex system digital Campus of the

UNESCO unitwin and um uh it is um university twinning network of 150 universities in the world and it would be great if UAH could join uh because again you’ve got a big big experience in complex system so it’s going to be really good um okay so let’s go back to

What complex systems are and how they were conceived how they were designed uh usually traditional Sciences um uh make a difference between hard sciences and soft Sciences okay and hard s Sciences would be applied sciences such as theoretical physics chemistry um and then uh you’ve got things that have are mixed between

Science and technique uh which is physics and computer science why is it different because in computer science you create the domain which is the science itself so this is why computer science is going really fast because you are creating the science as you develop it so it’s uh it’s going faster and then

You got things that are less hard such as experimental Sciences such as biology medicine uh typically it’s not because you have a high cholesterol that you will die the day the next day okay uh Churchill was known uh he um he died at an advanced age and he was really fit he

Had all his head until he was I don’t know more than 80 and journalists ask him okay what is your secret for being so uh good at 80 you know and Churchill said oh never sport never do any sport so that was his secret and he was he was

Having a bottle of whiskey every day okay and so you see this is less uh uh exact scientist it’s not because you you’re drinking a bottle of whiskey every day and you don’t do sport that you cannot go to Advanced age without having all all your head and then there

Are soft um Sciences which are typically more Humanities economy cognitive Sciences sociology and for people of heart Sciences they just look at at them you know from a higher point of view saying oh there are no numbers behind this no equations difficult to understand uh how all this is related

But in fact uh what we found out in the uh second half of the 20th century is that actually both could are exactly the same apart from mathematics I put mathematics here and you will see for a specific reason that will uh be unveiled here so the question is could hard and

Soft Sciences be similar or United and yes they can through complex systems and we’re going to see this together okay um ah so I’m going to go to the small tour so it will be this one okay so as I said it starts uh with a strange character here um um all this

Is the emanation of the Greek uh deep thinking of uh philosophy which is understanding um the world that is around us and it all started uh for Western science it all started with pines of aah and we’re going to talk uh about him in the second presentation

Later on okay and he is the guy who understood the importance of no of the notion of ontologies and ontologies represent intelligence they represent everything in the world and we’re going to talk uh more specifically about this on the second talk um then there has been uh different currents in the

Development of science there’s an very interesting current is which is the one of Lucifers democratus epicurus uh because in their view of science they said oh they understood that everything in the world is composed of atoms and void and they had the vision that atoms were actually turning uh there were in

Spinning constantly uh when atoms were um they could attach one to each other and when they attached they created a new kind of material that had different properties and the original atoms so they were already they understood the concept of molecules um then they had understood the concept of turbulence

They had understood this so absolutely incredible of course the notion of void was different than the notion of void of the 18th century which is also different than the the current notion of void where we have Quantum void where all particles are there and this is the

Notion of void is is evolving but it was already there U since 2,500 years ago uh unfortunately it is not their vision of the world that won it is another vision of the world that won it is a vision of Plato and Aristotle okay for other reasons I’m not going to get

Into this I could maybe in the questions and um they thought that no the world could not be composed of atoms and void um uh but there had to be dualism uh and where objects are the shadow of reality but fortunately um uh they had one Vision that was good through asral which

Was the whole is more than the sum of the parts and this is the vision of the world that is compatible with complex systems but apart from this unfortunately due to them those those two guys uh science stopped for 2,000 years sto developing and that was really

Really a big problem okay now science um the development of science started again uh typically by 1600 with jordano Bruno with uh um Galileo with uh tiob Copernicus and all these uh people um and it actually peaked with uh Newton and so the story is that Naple fell on

Uh Newton’s head and from here came out uh the discovery of the universal law of gravitation uh no I don’t want to upgrade my software at least not now okay um and from there uh came this famous equation and this is an a very important equation and you will see it

Will also be important in the fut in in the what comes next uh because it’s a very very simple equation it it says that the attraction force between uh two bodies is uh the constant of uh universal gravitation which is G times the mass of one body times the mass of a

Second body divided by the distance between the two squared so there are only one two three variables there and thanks to three variables only you could explain how uh the planets were rotating against the Sun so it was ah so they could predict eclipses you could predict

All things important at the time because astronomers um people were a lot into uh astrology and everything and it made people think that maybe the world could be all explained by exact equations just maybe not as simple as this but maybe larger equations maybe you’re going to

Missing things but the fact that such a small and simple equation could uh reflect the position how planets were running around the sun such important things uh was really very import very um um um had a lot of signification for for people but unfortunately I’m going to go

Directly to the 20th century in the 20th century uh many many things broke down and many certitudes that we have we were sure of things and the everything broke down and interestingly enough it started to break down with uh the three bodies gravitational problem uh which was a

Problem that wasn’t solved by Newton he understood that there were problems he said the question was for Newton it it said the the it established the value of the force for two bodies when there’s the sun and a planet by how much is the planet attracted to the Sun by how much

The sun is attracted to the planet the big question was okay but what if there are three bodies okay if there are maybe uh two suns and one planet or one planet and or One sun and two planets what is happening and what is going to be the um

Uh the trajectory right now with two bodies it’s an ellipse it’s very simple okay and that was an open question and in 88 9 point made a big discovery which is that uh the trajectories may not be elliptic anymore and we’re going to get into a very very strange behavior um

Which is called uh deterministic chaos and deterministic chaos is there I’m going to show you um with a small app so this is when you have two suns and may maybe one the Earth and so the Earth is going to do ellipses around the Sun but

It is attracted by some number one here and sometimes it is attracted so much that the planet will go the wrong direction of the Sun and then it will go this way and then maybe uh it’s going to go and visit some number one to see how

It what is it color and then it’s coming back and to the big surprise of Point car this is not an ellipse anymore and not only is it not an ellipse it is something that is uh not um a cycle meaning that never before has the planet

Been here in the past and never again with this planet come there exactly the same position in the future meaning that if we let this run for very long time the planet will just cover all the space in between there it will never ever come back into it’s not a cycle it’s not

Cyclic all right so uh what is here is called deterministic chaos it’s chaotic because you cannot find any uh reasonable way you say well it looks like this is doing ellipses but then at some point oh it’s going there nothing is is PR well you can predict it exactly

Because it is the equation of Newton with one more term which is the mass of the the other planet of the second of the sun number one here son number two it’s ex all still this very simple equation everything is exact and everything is exactly predictable if you know mathematically what is the

Weight of sun one what is the weight of sun 2 what is the weight of the planet what is the distance but you must know them exactly mathematically if you know things exactly everything is predictable but now we have this perfect equation do we know the exact mass of the

Earth if we if we get it wrong by one kilo the future may not be the same this is what he realized and it’s the same with the position between the Earth and the Sun or between the different Suns do we know the exact mass of the Sun so

Here we are we have we are with an a very simple equation that uh has a problem if we don’t know exactly the values of the mass of the distance then this is what we get I’m I did a a second uh simulation on using this app and here’s the simulation where um

We have so I’m going to pause it if I can yeah so here I started four different planets they don’t interact together the other planets they just four different runs okay where the planets did not start exactly at the same point let’s say that the planets started with one meter offset and when

They don’t start exactly at the same point you see that at the beginning the trajectories of the planet are still identical of course because uh it’s not because this table is not perfectly flat that it will that things will roll off it or it is roughly flat so that is good

Enough so here the trajectory to start with is still exactly the same then after a while the planet they their uh position changes slightly but it’s not really that much okay until um something happens that is really strange if I can’t manage to get there again to restart the thing uh

Yeah okay and so the four planets they follow roughly the same um uh position same trajectory unless we get on the wrong side of the Sun and here what the planets are doing well you will see that their trajectories is not exactly the same even though they started at nearly

The very same initial point and you see that now their trajectories are all over the place okay meaning um that with Newton’s equation we could predict the position of the planets until a certain point and after this point which is called the U uh Horizon of uh uh sorry I’m really bad with

Names um it is the Horizon anyone in who knows in complex systems it is liapunov’s Horizon okay um so liapunov’s Horizon means until when until when we can make a prediction and Beyond uh leonov’s Horizon then Things become to be completely uh chaotic so this is what

You get even though all the planets started just nearly at the same place so this means that now um we have equations but if we don’t know exactly the values of the variables in this equation we can’t do anything about them because this is what we get and this was a huge

Comprehension new comprehens of of complexity and so what you see here is Newton’s own equation but applied to three bodies very simple equation in chaotic system sensitivity to initial conditions makes it impossible to predict the future all right and this thing was understood uh by Point car in 1989 but

He he was a mathematician and the time for uh scientific knowledge to go from mathematics to other domains of science it takes time also and it was not until uh 1972 that climatologists uh or maybe just a bit before they they saw some really strange things happening when

They tried to to predict um where hurricanes uh would would appear okay they had a model of the sea they had a model of everything and then they changed a little value on the temperature of the sea and in a certain space and they run the simulation again

And suddenly you had hurricanes at different places and they didn’t understand why they just changed the value of the Sea of the temperature by uh one10 of a degree so they would expect okay maybe the hurricane would not be at this position it would be just

A bit away but it the the the future were completely different and so this is the origin of the um uh butterflies uh effect that everybody now knows about so the Lawrence made a public conference on does a flap of butterflies Wing in Brazil set off a

Tornado in Texas is it possible because when a a a butterfly flaps its wing in Brazil or possibly in Chile there could be a small swirl that comes off the The Wing of the butterfly and this will change the future forever because the initial conditions are not the same

Whether the butterfly did not flap its wing or whether the butterfly flapped its wing the initial mathematic the conditions are not the same therefore the future can be completely different uh just for everyone it’s not completely true because uh it is a damped uh system the atmosphere so butterflies can flap

Their wings it’s not going to change the climate but it’s just for the image all right that um uh with complex systems a change in the initials condition means that you don’t you cannot predict the future exactly anymore and this is bad for exact equations because when you had

Exact equations you had the hope of being able to understand and to to say what would happen exactly because the equation is exact but if you don’t have the exact values to put into the equations then you don’t know all right um and then in the 20th century many terrible things happened to

Science for examp even in mathematics G incompleteness theorems uh just broke down arithmetics um and then um uh and then there’s another interesting guy uh who is friend and um he was also a professor in striborg University uh he got a Fields medal which is the equivalent of the

Noble priz in mathematics in 1958 uh but in 68 he came up with a theory which is called the theory of catastrophes okay and what is the catastrophe theory that is something that he understood he was a pure topological guy nothing applied a complete uh fundamental mathematician

Okay and what he just realized is the following it is that in a space up to four dimensions there are only seven different ways to go from a stable state to another stable State not six not not eight not five okay just exactly seven so what it means to go from a stable

Topological state to another topological state I’m just going to give you an example with the most famous bation that exists which is bation number two it’s called the cusp catastrophe um the there are many physical interpret well there are interpretation of this everywhere uh an easy one could be that

Suppose that you have a table and on the table you’ve got a napkin okay and the napkin has a fold here and then you’ve got maybe a marble you’ve got an object there and you see that this object you can push it around and it will be able

To come back to its original place in a stable way uh because it’s there if you push the ball there then it will stay there if you push it back it will stay there not nothing happens and you can always get back but if you go a bit uh

Um um too far in this direction then there is the possibility that the ball Falls and once it has fallen down something has happened it’s called a catastrophe and then from point A2 can you go back to point A1 by the same way as you went from A1 to A2 no because

When you’re going to go back well the ball will is going to get stuck into into the fold here so there’s no way until you jumped back you know but this wouldn’t be a a stable uh going from a stable state to another stable you can’t

Go there back but it’s not true there is a way to come back to A1 through stable states which is by uh going around the fold if you go around the fold from here you’re in stable State you can go there which is another stable state which you

Can go there which so it can go from A2 to A1 by using only stable States if you go around it okay and this has applications everywhere suppose that you’re doing sports okay and sometimes you if you are putting too much stress on a bone then the bone will

Of course Flex like a piece of wood all right so it will Flex this way and then when you release the pressure the bone will take back its uh its shape and uh you can do this okay can when you’re doing sports you’re putting stress on

Your bones but if at some point you put too much stress on the bone then what happens some well the bone May Crack and then you’ve got a cracked bone and when you’ve got a cracked bone H you could try to put the two pieces of of the bone

Uh in front of the other again but you will not get a a unique bone anymore you still have two pieces that are cracked they’re back into position but they’re not glued anymore so you could say that well you can’t go back to the original state of

The bone unbroken but it’s not true because if you weigh for uh 4 weeks then uh without moving then you will have um uh particles of bone that will uh be recreated by the blood and everything and the bone will again become um uh unbroken and you can start again if you

Like it you can start breaking your bone again but what this means is that you see that this scheme here this uh small um um uh uh image shows what happens to people when they become sick you get uh I don’t know you you are in good health

You’re in a stable State and then you get covid or you get the flu or you get something and then suddenly um you become into into another stable State uh which is that you have a rainy nose you have temperature you’re not okay can you

Get back the same way um you become you became ill can you get back to health the same way no you can’t you have to wait for your immunity uh system to heal you and then later on you can hopefully get back to the original state again so this has

Applications in it has applications in absolutely all domains all right knowing that there is not only the cusp which is the second uh of all the um uh the bations and the catastrophes but there are seven of them okay there the fall the CP the uh and um the hyperic umbilic

The uh elliptic umbilic and what’s interesting is that they all have uh a graphical representation but they also have a spatial interpretation through a noun um and they have a temporal interpretation which would be a verb and the temporal implementation has two variants one variant is the destructive

Interpretation the other variant is the constructive interpretation so let’s take for instance the tical umbilic all right um You can see that this could be um a needle or a pointed thing and with this needle you can make a hole that would be a destructive uh um uh

Operation by using the elliptic umbilic you have a a napkin you can punch a hole into the napkin but with the same um uh metaphor here you can also put it upside down and you can plug a bottle and so you can plug a hole and so this would be

The destructive uh effect that would be the destructive effect and all the different uh um catastrophes have their own signification okay now the big thing is that for instance in aeda medicine aova is a uh the traditional Medicine of India uh they have a vision of a cold

Which is a hyperbolic on umbilic which is that when you’re in good health you’re on top of the crest here and um they say that if you get the cold you’re going to be into a wet State why are you into a wet State because suddenly your

Nose is going to run your uh eyes were going to be to become wet you’re going to to have temperature you’re going to sweat so you’re in a wet state so you’re going into this side of of the um of the Hill which is the wet side of the hill

And as a cure for your um cold they said oh you should stop uh drinking you should you should try to or reduce it and you should eat um uh things that are dry because you are into a wet State and you have to get back into your original

State of dryness so you should concentrate in eating bread and eating things that are that don’t have humidity into them in in the hope of of healing them and then miraculously after one week people are healed but maybe not for the good reasons so that’s another view so the

View of the world maybe here if you’re on this side how can you go to the other side of the elliptic umbilic can can you go up to the top and then come down but this is infinite or you have to circle around you so this gives you an idea of

The topology of how to get back to the original state after a catastrophe has happened and when you’re a scientist if you know to which catastrophe the problem belongs then this gives you ideas on how to solve the problem because again in the world we are in a

Four-dimensional world there’s x y z plus time so there is no other way to go from one state to another than these seven ways good now um yeah so applications in math epistemology biology morphan Genesis Health Humanities it’s everywhere because we are in a four-dimensional world um now let’s go

Back to to uh philosophy and let’s go back to the difference between uh mathematics and and the rest of the Sciences because you show when I showed the the thing at the beginning I said well mathematics is different and here is a real reason why mathematics is

Different um in 1905 finally Jean peran a french guy demonstrated that atoms exist okay so now from 1905 we know that uh the room the objects the chairs everything is and ourselves are made of atoms and void so we are back to the vision of Lucius of the origin of 25 500

Years ago but this has a huge implication the fact that everything around us is made of atoms um the big question now is that uh ukus 20300 years ago demonstrated that there are only five uh regular convex poly hedra what is a regular convex polyhedron um Regular means that it is

Made of always the same um uh surfaces okay here’s triangles and here you’ve got squares here you’ve got pentagons convex means that it’s a solid that doesn’t have a hole in it it’s just uh you you it’s inflated okay and polyhedra is made of different uh different shapes

And here again as for catastrophes you don’t have seven you don’t have six or you I have only five of them with a pure mathematical demonstration by ukus 2,300 years ago so here are the they are called the platonic solids but it has a huge implication the huge implication is

That now suppose we want to create a sphere that is made of atoms do perfect spheres exist if perfect if a perfect sphere was made of atom so you say that um uh the each individual object would be an atom of silicon of anything and if it’s perfect

Sphere then it has to be convex there cannot be any hole in it um and if it’s make of all the same atoms then it has to be regular now um is a sphere like this part of these five platonic solid is not there so what is a sphere and canos

Sphere exist mathematically uh so you say yes of course because you see it we can do with hexagons we can put many many hexagons together and it will create a sphere but it’s not exactly true because if I zoom on this sphere you will see something strange here

You’ve got a pentagon what is it doing here you got hexagons everywhere here’s that’s not a hexagon um and if I unzoom a bit on it um ah there’s another Pentagon here uh oh there’s a pentagon there so you see there are pentagons here and there what

Are they doing into this perfect sphere it is not well in fact mically speaking there is no uh tiling of a sphere that can be done it doesn’t not exist mathematics say that there is no tying of sphere that is possible if it’s made

Of a small of number of of of of of entities now if we put all this together we get into a a strange uh conclusion if we put this back into a bation now suppose I take a mathematician guy okay and I have a surface here that is

Perfectly um um symmetrical so we can have an equation for this that is completely U uh a symmetrical equation between uh if that is the value zero of this angle of this symmetry plane the the the area is absolutely the same if I take a mathematician I told him I tell

Him okay here is a sphere a perfect sphere what happens if you throw a perfect sphere through through this Valley before it h it goes onto the hill there if the guy is a mathematician the sphere he will say we go along the ridge of the of the Hill because it it’s a

It’s a perfect sphere the uh landscape is perfectly symmetrical the sphere has no reason to go left or it has no reason to go right so now I say okay let’s I just printed this perfect uh thing for you with a 3D printer and well you should try with something that I have

Which is a sphere which is a marble and please try it with a marble and when he tries it with a marble what will happen is that of course the marble will go left the marble will go right and if the mathematician did not really understand

His uh um his uh uh domain he will say oh yes but it’s because the marble I I didn’t manage to throw it exactly at the same at at the right angle on the right direction but now you understood it’s not true uh the marble here will never

Go up there because the marble is not a sphere and cannot be a sphere mathematically there will always be an atom that is too much on the right or an atom that is too much on on the left because perfect spheres cannot be tied mathematically so there is a mathematical demonstration that says

That oh there is going to be a bation here and this is how you relate um physics to bations you cannot Escape bations when you’re using physics because physics is made of atoms not and no exact equation can represent a sphere can represent anything

So um now we are back into a big problem which is that um we had Newtonian physics before that was a top- down representation of the world through equations and we see that just because of a sphere on a marble it just simply cannot explain the world so we have to

To have a new understanding of how the world is working and this understanding came um thanks toare but it came into the second middle of the 20th century it’s very recent and basically it was in the 1970s that people understood all this so you see science evolved for many

Many hundreds of years until it was completely broken down in the 1970s and now we have to define a new way to represent science and this new way to represent science is this the science of complex systems it’s a new science that takes and you will see takes everything

Together so the definition is the following a complex system is a system made of a large number of autonomous entities in interaction that can create uh several levels of collective organization leading to emergent and IM IM mergent Behavior it’s a very long sentence but it describes all Sciences

So it can be 10 words long um now I’m going to explain this uh difficult sentence to you um now a complex system which is anything that is around us is made and let’s take this uh Sphere for instance this marble uh it’s made of a large number of autonomous entities the

Autonomous entities would be the atoms that create the marble um they are in interaction because the atoms there are some valence forces that hold the atoms together um and they create several levels of collective organization leading to emergent and emergent behavior for the beginning and just going to uh uh concentrate on emergent

Behavior Uh if I take one billion atoms and I put them together into the rough shape of a sphere what will happen is that I will have a marble and the emerg what is the emergent behavior of a marble do you have an idea what the emergent behavior of a marble is I

Usually have a marble in me but I just forgot to take it it’s still um in my room what is the emerin behavior of a marble have you seen a marble and have you played with a marble when you were young well if you take a

Marble and if you put it on the table and you push it the marble will roll it will roll in any direction it will roll not perfectly because it’s not a sphere but it will roll relatively well so that’s the emerging behavior of of marble now if you take the same billion

Atoms and you organize them not in a marble but in a cube if you take a cube and you push the cube on the table the cube not roll so you see a cube and so that would be the first level of collective organization a marble rolls a

Cube made of the same atoms does not roll then the second level of because has several levels of collective of imagin behavior it is that when you take 10 cubes and you give this to a three-year-old the first thing that the three-year-old will do he will put the

Cubes one on the other and with the 10 cubes he can create a wall okay now I say okay oh this great you created a wall the 10 cubes now here is a a bag with 10 marbles can you create a wall with with 10 marbles you will put the first ball it

Will stay there second ball the first one uh of course it will fall there is no way to make a wall with 10 marbles even so that would be the second level of of emergent Behavior first level cubes they don’t roll second level oh with cubes you can make houses but if

Only had marbles you just cannot build anything out of this that’s the second level up there emergent Behavior would be what happens for instance if you get a virus and a virus is really a tiny teeny piece of uh uh proteins okay and when you ingest the virus suddenly you

Were a pure healthy uh animal and suddenly you become ill so that would be emergence it would it would be U uh what a really tiny so emergence means it is a bottom up behavior all the atoms together will make a ball to roll will

Make a house to be built with brakes and everything the emerging behavior is the opposite is how a tiny little B piece of thing will can infect the whole body and put it into a different state into another bation right so what we now have is a new comp in the new science

Which is the science of complex systems it is a post Newtonian Paradigm that is deeply rooted in the fundamental laws of physics and Mathematics of course and now it led IIA pin who was Nobel Prize of physics of 1977 who to say the following sentence that is extremely

Important that since we understood all this the aim of science is not to predict what will happen because we cannot know what will happen exactly in the future it’s impossible it’s finished we should forget about this now the only thing we can do is to try to predict

What could happen with a probability the the planets in the future if in the near future oh yes they will be here the planet longer future they could be here and then after the liapunov Horizon they can be anywhere and we have no idea where it can be because we don’t know

The exact mass of the planet we don’t know the exact distance between the planet and the sun we don’t know the exact mass of the sun we don’t know anything for sure exactly mathematically and as soon as there’s a small difference Butterfly Effect poof the future will will be

Different okay so now where do we have complex systems uh around us um well and this is a great example of emergence you have seen grains of sand and when grains of sand interact through water or through air oh well they will create small Dunes like this and interestingly enough that was in

2022 last year uh in nature was a a paper on Mega rle mechanics B bodal Transport ingrain in bimodal Sands and the contents of this paper is that and you see here you’ve got two um two levels of emergence you’ve got the small ridges small um uh undulations here

Waves and the bigger waves which are big tunes and this Bodel um um uh transport uh they found the equations of it if there are equations behind uh the ripples there and behind the Dune well it means that when many grains of sand are put to interact together oh they

Make mathematics which is not bad because a grain of sand is not very intelligent okay so on its own it doesn’t have many neurons doesn’t have many things too but when you put millions of them together oh they they make M mass and we found their equations

That they are building together so you see a beginning of a comes com from kind of intelligence that comes from things that are absolutely and totally non intelligent there again there’s nothing less intelligent than a grain of sand but together they do maths then what you have is though if you have

Cognition uh what is cognition made of and how does it occur well in in our brains we have around 100 billion neurons bit less 80 billion neurons and and um when we take an action where does this action comes from it comes from the interaction of these billions of neurons

And in one ten of a second if there’s a a car coming coming towards us a bus you know we’re going to jump somewhere to avoid it uh the the decision is taken one10 it may not be the good decision but actually with the billion neurons the hundred the the the thousand billion

Neurons will take a decision and you will do something and you all know that if you have been in a meeting if you have more than 10 people in a meeting it’s going to take time to take a decisions they do it in 100 in a tenth

Of a second all together okay and so you see that this is an emerging process of a complex systems we have billions of not intelligent neurons when you put them all together you’ve got intelligence at the end um in biology well what are animals made of well they

Are made of cells and the cells would be the individual The Entity and when you put many cells together you get organs and the organs together when you put them all together you get an animal and if the entities are interacting you get a living animal if you uh stop the

Interaction in between the entities well you get a fish at the shop and the fish is not living anymore it is still a fish it contains all the cells of a fish all all the organs but interaction is lost so you see that here not only intelligence emerges from the

Interaction but here it’s life that emerges from the interaction so as soon as the entities don’t interact anymore the fish is dead um but that is a single animal but then what you can do so you see we went from physics to with grains of sand and

Of course um with a climate you this is the Navia Stokes equations of turbulence um but when you have many animals together there The Entity can be an ant and when you put hundreds of ants together they can create Bridges they can find the smallest uh path between

Point a and point B they become they increase their level of intelligence in an emergent way um and then when it works not only with ants uh but also with people it’s called a social network and when it’s a social network it’s also many entities that are interacting

Together in order to create emergent behavior and you can even have this in a virtual way and what is the virtual way uh well that will be internet and the web the web is made of billions of pages all in all pages you’ve got a link towards another page and then you’ve got

Amazon that is emerging from it or you’ve got Tik Tok or you’ve got emergencies that everyone uses and all these things ah are following exactly the same principle so we’re not into exact math into exact physics into exact chemistry but now all this previous science the old science was toped down

With equations that tell oh this will do this and that and now we understand understand that we cannot model this anymore in a top down way you have to model it from a bottom up way and as soon as you model things from a bottom

Up way uh well sociology is part of it and so we reunited all different Sciences the hard and the soft Sciences are all the same because they are made of bottom up uh Behavior okay so let’s have a look at uh different uh domains ah yes something

Very important ah uh please um Constanta or uh someone I need I need to have a pens could you please fetch me a pen and an eraser I need this here okay so um can we understand um uh the emerging Behavior it’s important to understand it mathematically speaking

Okay because uh um I was as uh uh Christo said I was a head of the the department of computer science in Strasburg University one of big universities and next to me was a head of the department of mathematics and when I told him oh yes but you know the

Whole is more than the sum of the parts it means that 1+ 1 is more than two he just looked at me and said ah another uh another computer scientist doesn’t know anything in mathematics because 1 plus one is two it cannot be more than two

That doesn’t make any sense and and he was right because um H would you have another color two colors red or that would sorry I didn’t say it um thank you and um uh and he was right because in mathematics which is the science a different science which is an exact

Science as we saw with spheres uh 1+ one is in fact exactly equal to two but as soon as you have a science which is emergent which relies on interaction um uh Aristotle was was right and it is possible that the whole can be more than the sum of the part so

Let’s try to understand it mathematically rigorously can we show this mathematically and interestingly enough we can so and this is the transition between mass and physics it’s it’s really beautiful so first of all uh just the uh intuition um of why the whole could be more than some

Of the parts is the following suppose that you are in in front of a desert and you want to go from point this point to another town or to the sea that is 100 kilometers away on the other side of the desert uh what you are going to have is

That either you can go on foot but it’s going to be difficult so what you could do is that you could take an SUV and uh of course and hope to to to cross the desert but if the SUV doesn’t have big enough wheels or if the sand is too soft

What you could end up is you could end up being just completely stranded as this guy and he has created such a hole that the whole car is in the hole and now the only way out for this car is just to be able to remove the Dune in

Front of it and going to be a big a big job um now you could say okay uh so maybe he will get stuck at around I don’t know three or four kilometers and the distance you can go without getting stuck is roughly um a ratio between the

Power of the SUV and the size of its wheels and the softness of the of the sand and if he could do it in 3 or 4 kilometers you can try it again maybe you will get to five but at some point the ratio will be the same for this kind

Of car and this kind of sand sand and you will get stuck so you could try to say okay let’s go with three or four or 10 cars together to see if we can do the 100 kilm and there’s no reason roughly people will be stuck at around 5 kilm

For the but everything changes if you get an interaction between the entities and the interaction would be okay now let’s have let’s say that all SUVs have a cable and they have a winch and if there’s an SUV that gos into a hole then what can happen is that the second SUV

That’s behind um can throw its cable and with the winch he can just uh take it out of the winch and then the SUV can go around okay and maybe hope to do another 4 kilm and by the first one um uh the second one helping out the first one and

The first one helping out the second one then by helping together by interacting you could hope to be able to cross the the desert but of course if the two get into the same big hole then they are going to be stuck so that’s nice that’s

Why it’s nice to be three maybe four uh so that you minimize the chance to be all to get stuck in the same hole but this is still not very mathematical it’s just more in an intuitive way to understand how interacting together will get you the emergent behavior of being

Able to cross 100 kilometers that you can’t do without it now here is the interesting juicy part and it comes from jeul De who is both a mathematician and um um computer scientist and he sort of in computer science we use a metric which is called benis uh logical depth okay

And it computes somehow uh I’m going to explain it in more uh intuitive terms it computes the complexity of an object an image a piece of information meaning does it have a structure in it and if the thing has zero structure meaning for for instance if this image is completely random thank

You thank you very much um if this is completely random it’s complex it’s uh the the Benet logical depth of this image will be zero because there is no structure you cannot find anything in it so you can measure its entropy is maximum at all levels okay this is a

Completely random image what happens if you take a completely random image and you combine it with another completely random image then normally if this is random and you combine it with another random image what you should get you should get another random image because random plus random is completely random

But if you take these two exact images and you combine them together with an exclusive or operator what you get is the head of Napoleon how is it possible to get something out of nothing okay and it comes from a uh I didn’t do it uh here

Digitally uh but I can do it very easily for you to explain how it works so then not going to to do it with a a big complex image uh we can do it let’s say with a u a 4×4 image okay so if it’s a 4×4 image it

Will have 16 pixels and let’s say that um we have a random image uh first a completely random image so I’m going to take I don’t know this pixel B to be black this other one to be black this other one to be black so I need if I

Want to be completely random there should be eight8 1 2 3 four I’m making them complet I’m trying to make them completely random I will not manage to do it so I don’t know this one so 1 2 3 4 five six uh seven okay and eight so I tried to make

A completely random image and now let’s suppose that I have another image this one that is not not random at all so I’m going to have another image of um and this image let’s make it that all the pixels here on the left are black so all these pixels are black okay

And all the rest are completely white now what happens if I combine these two images with an exclusive or so this is a bit for computer scientists but it’s not very uh difficult to understand what will happen there so an exclusive all means that it

Will be the result of one or X or one will be one if it’s um if the result of a X or B will be one if a is one and B is zero or if a is zero and B is one if

A is one and B is one you’ve got zero if a is zero and B is zero you get zero so this is what the meaning of is an exclusive or is and so let’s have a look at what we have here so the first it it

Must be exclusive you cannot be if both at the same state then you get a zero if both are at a different state then you get a one so here you got a one and a one so this will be a zero here here we’ve got a zero and a one ah they’re

Different so it means that this here will have um um black uh black pixel now here zero is zero is zero okay zero is z is zero and here again now I have one and one because they are in the same state it means it’s zero zero and one Ah

That’s a different state so this is one okay uh one and zero oh this is one okay and one and zero oh this is one now third row zero and one it is one okay one and one is zero 0 and 0 is 0 0

And here I have one and zero so it is one okay one and zero it is one again um zero and zero is zero and one and zero is one so this is the image I get and now what is this image made off let’s and this here that I will have the

Second color thank you um you will see that that on the part where all the pixels were black on this area here which is this area of the combined image what you get you get the opposite of what was there because when there was

A one if there is a one you get a zero if there was a one and you get if here there was a zero then you get a one so you see that here what is there is the exact opposite of what you had here okay

Do you do you see all this wherever there was a one you zero there’s a zero there was a one it’s the same so if originally this part was random so this has to be also random but it’s an inverted random it is also random but where there was a one there’s

A zero but it’s also random and on the second part what do we get on the second part we get that um oh look on this part here it is exactly the same because whenever there was a zero there is a zero when there’s a one

There’s a one and so this is exactly the same so if this one was random then this one is also random because it is the original one so you’ve got this image which is random where there was a white because it’s the original image and when there

Are black pixels it is random but reverse but it’s also random and now the miracle is going to happen um suppose now that I’m erasing this one so if I’m erasing this image what do I get I get two different random images completely random there are these two images there and what

Happens if I try to combine the two random images so this is the uh benit logical depth of zero zero complexity benit logical depth of zero and um so this one wasn’t working well so here now let’s recombine the two okay so this are my 4×4 image pixels and let’s do an

XR between the two if they are identical I’ve got a zero if they are different I’ve got a one one zero they’re different so I’ve got a one 0 one oh they’re different so I’ve got a one 0 0 oh they’re identical so I’ve got

A zero 0 Z identical I’ve got a zero okay now let’s have a look at the second row one and zero are are different because they’re different I’ve got a one and you see what is coming up zero and one oh they’re different so I’ve got a

One um one and one oh they’re the same so it’s a zero one and one they’re the same so it’s a zero so third row Z and zero are uh ah sorry zero and one they’re different so I get a one okay one and zero they’re different

So I get a one then the rest zero they’re identical so I get zeros and you get it and so the result for my uh colleague who was head of the uh Department of uh mathematics is that out of two images that are Rand completely random of complexity of uh uh

Benet logical depth zero when you associate two completely random images you get an image that is not random at all so you see that the whole is more than the sum of the parts and why is it possible it’s because the two different entities were in interaction how were

They were in interaction this one has been created out of that one and this one it’s a combination of the two that created this one but this one is also random so when you remove them the two are in independent entities independent random entities but they are in connection because they have been

Created by this one and when you recombine the two you get back the first one that is not random at all so this is the mathematical demonstration of emergence okay so you see that whole is more than the sum of the parts and uh computer scientists are not completely

Stupid when they say that 1+ one can be more than two and there is more it’s possible to recombine things in certain way that you get more than what you get there so the complexity of a union B is greater than the complexity of a which

Is zero plus the complexity of B which is zero that’s the complexity of a union B it’s the head of Napoleon it’s not it’s not random okay okay so uh one hour that’s perfect uh so I still have yeah I will try to get uh time for questions so now let’s

Have a look at uh where complex systems are around us I already showed a couple of uh places but let’s go let’s look more into details well in physics there is a domain which is called fluid Dynamics um and in fluid dynamics this is what happens if you uh look if you

Put some color into water that is Flowing then you will get turbulence here all right and what happens is that in the 197 in the 1880s or I think um two guys a french guy and an american guy uh found the equation behind this okay the French guy was neier the

American guy was Stokes and both of us now got Navia stes equations and with Navia Stokes equations this is what we can do into a computer so we’re going to have if I manage to click correctly this is the flow and these are NV Stokes equations and you will have a simulation

Here of the turbulence now is this going to be exactly the same simulation as what is there no because again we cannot know ex the exact initial state so the future will be different but it will be turbulent the same way and in order to make sure that it is turbulent the same

Way we can uh have a look at a simulation of a higher um emergent level there are again several levels of emergence first level of emergence is you’ve got a turbulant flow second level of emergence uh it is called uh font cat for C Vortex shedding so here you’ve got the flow

That is going from right to the left and when you put a cylinder there and you put red ink on the top and blue ink on the bottom you will see something very strange to happen which is here you will have a red swirl

And a blue swirl and a red swirl and a blue swirl and you get this alternative uh Behavior which is a higher level of emergence and what happens if we take the Navia STS equations and we put them into a simulator of nav St equation this is what we get so you will

See here you see the red and blue Swirls and you get the same red red and blue swirl so it means that the model of mathematical model we have can explain a higher level uh of just turbulence it is more than this and what is nice is that uh when you go

Into uh a higher scale not a higher level with a higher scale uh this is what happens with the um island of Guadalupe on the the west of Mexico and um when you’ve got the right uh uh clouds with a satellite image what you see here is that you’ve got the

Fontan uh um vortices that are appearing so not only it appears into a small tube but it appears at the scale of an island when they’ there’s the wind that is uh correctly that comes correctly and and uh um humidity that creates uh uh clouds but then what happens is that if you

Look at the at the scale of the Earth this is taken from a satellite umat okay okay and you see that you get some turbulences here that can evolve into possibly this one you will evolve into a hurricane that will hit Florida okay then they go from Africa and this is how

They evolve and if you’ve got a good enough model of the Earth into a computer well this is what you get and you get the same thing okay uh now the coast of africas are here and it’s a bit accelerated but you get some turbulence that is starting

There and that is going to hit uh the coast of uh um of America here and Florida there and you’ve also got some uh hurricanes that go back to Europe and it happens once in a while it happened uh uh last week or two weeks ago in uh

France and in Germany we had a t we had a hurricane that went back towards the coast of uh Spain France and England and Germany so we have SE and not only it is at the at the scale of of the Earth but you can go at a higher scale and a

Higher scale this is the scale of galaxies and here are two galaxies that are going one one into each other and when you apply uh Navia Stokes to the galaxies this is what you get this is a simulation you’ve got the two galaxies that are interacting and you get this

And why is it interesting it is because if you look with a telescope at the galaxies if I can start this again okay then sometimes you will see in the sky you will see this shape and the fact that you see this shape in the sky you

Know that it is because you had two elliptic galaxies that went one to each other and you CAU them at this stage and then you can see also you can see them at at that stage and you see something with two nodes and some arms around here

So it means it’s a collision of Galaxy and it is the same equation and the one that is water so it is a complex systems entities that are interacting together through gravitation okay um so this is for physics this is really deep physics and now in chemistry

Um uh now we can so breing at time that’s good so when you have molecules uh molecules can interact together and this is how you get a water molecule so you’ve got two hydrogen gas molecules okay and then you’ve got a an oxygen gas molecule and when the energy level is

Low enough you you you get water molecules and when there’s not too much heat then you’ve got a hydrogen links the hydrogen link that happens between the molecule and when you’ve got the hydrogen link uh then it means that uh two uh gas molecules can be linked

Together and the hydrogen bond here is what will create the liquid out of the gas okay and then you’ve got these two that are here and as the molecules are bonding together the gas is going to become more and more liquid and you’ve got a phase transition and the phase

Transition is a catastrophe um uh and suddenly you go between gas to liquid and or between liquid to Crystal and it is exactly the same thing that is happening through a cusp uh catastrophe oops sorry for that okay and from there on you can simulate complex systems in a bottom way

Up so here is DNA replication okay from the bottom way up so they managed to simulate all the um all the all the the cells no it’s not even cells it’s within cells okay that’s the um the small replication machine that that duplicates DNA and you can do it again from a bottom

From bottom work way so this is would be in uh in in chemistry in biological chemistry okay and then if you go from biology you can go ah from the cell to the organ that’s going to bother me sorry because there’s sound there okay now right so this is something that has been

Done by the University of Tokyo they have reconstituted a heart um from all the cells that compose a heart and when you put back the all the cells have been reassembled at their correct uh place and and when they get to interact together well they’ve got a beating

Heart and you can compute how much blood the heart is uh is pumping through and all the cells are there and they’re put of course no two hearts are identical because of the original um uh initial state is going to be different so all hearts are going to

Behave a little bit differently but generally speaking they will pump the same way and so you can do U model simulations of uh of organs um and then this is something that we’ve done in my team uh not completely we’ve done the computation on a sub super computer of

My team here is embryo Genesis um of a uh um uh uh a zebra fish and what is nice with a zebra fish it is that it is transparent and it is small enough so that it can go into a laser microscope and into the laser M microscope what you

See here is that you’ve got the development of the embryo if I can get this to be here uh so let me start again there so you see it starts with two cells and the two cells divide when the two cells divide you’ve got four cells

Starts with one then two then four and we and then the four cells each divide into two so we’ve got eight cells okay and the eight cells will divide into two so we’ve got 16 cells and you see them really clearly and you see that we already have a a

Mathematical model of the development of an embryo it is 1 2 4 8 16 but and all these cells are identical this is why they’re called stem cells you can create um brain neurons or bone or any kind of tissue out of these cells because they have not differentiated yet but then you

Will see a moment of differentiation that comes there which is when all the cells cannot be in one layer on the yolk this is the energy that’s the yellow of the egg okay this where it gets its energy from and at some points you will

See that the cells has to go into two layers so here it was still in one layer and suddenly you will get a plop and it will happen there then there it happened and you see that the the cells will go from one layer to two layers I will try to

Get it there and suddenly you will see a plop and uh no I went to I didn’t so it’s here all cells are in one layer and suddenly you will see plop they will plop into two layers and when this happens there’s a big problem for the

Cells that the second layer the cells of the first layer they still have access to the energy directly to the Yol but the ones that are on the second layer H they don’t have access to the yolk anymore they have to ask the first ones

Hey can you I need a bit of nutrients can you pass it me through so for these cells that are on the second layer the initial conditions have changed and if the initial conditions have changed their future will change and they’ve got the DNA which tells them what how how

They can um um differentiate and so this is what we’ve done in my team uh a PhD student uh and so he has done the same thing in a computer so from 2 to 4 to 8 cells and so on and then at some point here you saw it you have the differentiation

Between in two layers and then uh we’ve got the blastoderms that go down and we have oh yes and this is important we have the same development as a zebra fish that is there if I manage to get it there because on a zebra fish the head

Will develop on the left the tail will develop on the right and we’ve got the same difference ation that that happens there now what is nice with Zepp fish is we know which part is the head we know which part is the tail so when we have

A a big amount of cells that are of the same type we know oh this is the liver this is the heart and we know because we have it for the real real thing and as soon as we know that this is the liver oh well we can replace all these cells

By a mathematical model of the organ you saw that we we did it with a human heart so if these cells will become the heart of the zebra fish we can replace these cells with a mathematical model of the heart and when you get all the different mathematical models there are partial

Models together well the idea is that you can you can recompose the animal completely in a computer and so there’s a big big number of teams on complex system that are working on this and um there’s a hope that by 2040 we will have model all the different submodels enough so that we

Can recreate the development of a human inside a computer from the stem cells to the adult stage to the uh to the old and death stage um and what is will this allow us to do it will allow us to make to create fory Health now this is

An important slide because um there are some building principles behind complex systems and all complex systems are what is called 4p and here are the 4p that is a thing that we organized in 2012 you see that’s a long time ago in Strasburg it says simul digital simulation for

Health from the cell to a virtual human being okay uh numerical SS to Virtual patient and right now we are trying to do a virtual patient or uh with a Mediterranean we’re trying to do um um uh a digital twin of things you’ve heard digital twins this is what we’re doing

This is the real uh human being and this is its digital twin and so when a baby will will get born he will have a set of equations and then he will develop and as he get the flu as he get ill as he breaks an arm we can make the same

Modifications on the digital twin of the individual and then supposingly he gets cancer uh at 50 the idea is to be able to test a molecule on the digital twin to see if the um I don’t know if if the kidney system can take the molecule if

Uh and we can test things on the gital tool before it we we do it on the on on on the real person so why is this 4 P um it is 4p for the following reason uh all complex systems are participative predictive preventive and they can be personalized so apply to

Health it’s a good idea a good example uh applied to women uh it happens so that uh large number of women get breast cancer so because they get breast cancer there has been a big um study that was made and we have been asking uh doctors have me asking women to participate to

This to so they created a cohort and the cohort is collection of data through participation you get all the different points of data and then for women who got breast cancer um the DNA was analyzed and in the end so you have a trajectory for breast cancer or for

Health and you will see that women with brca2 u uh mutation on their genes brca2 means breast cancer they will have an 80% chance of developing breast cancer when they are after the age of 35 or 40 so once now you take a new woman uh who uh whose mother

Aunt have had breast cancer at a early age well what you can do is that you can do a DNA test and if on the DNA test you see the brca2 mutation well you can now predict using the trajectories you say oh unfortunately you have the brc2 mutation

So it is very probable or it is you can say will with the probability of the cohort here you can say oh well it is probably at 80% that you could develop a breast cancer or an ovary cancer and this happened to a very famous woman um

Whose name is uh she’s an actress uh I’m I’m very bad with name so I’m just forgetting her name um it is um uh Angel Angelina joli and Angelina joli some six years ago now her mother died of breast cancer her a died of ovary cancer or

Maybe the opposite and so she made a test and unfortunately she had uh this mutation and because she had this mutation so was predicted that she could very probably get breast cancer so then what happens is that now that we’ve got the model a mathematical model out of

The participation now we can do predictions using the model and now what we can do is that we can prevent catastrophes and the catastrophe for Angeli one year she had her breasts removed and reconstructed but removed and then the next year she had her ovaries removed in order to try to do

This so that breast cancer doesn’t occur with her if you remove the breast and you redo then do you minimize the Chan of breast cancer developing um and what’s important that you can do this in a personalized way because you have the mathematical model of the patient so

You’re going to see what kind of action the patient can tolerate best and so this is completely a part of complex systems okay they are all about this thing once you have the mathematical model out of the data you can use him to pre to pred predict and if what you

Predict is a catastrophe then you’re trying to do okay what can I change into the initial conditions which are the conditions the current conditions so that the catastrophe doesn’t happen and then we can do it um if if it’s a floods that you’re going to predict then you’re

Trying to see in this special configuration of the uh collecting water uh then where should can can we create some basins so as to pre prevent floods from occurring and everything okay now social behavior this is the last um uh example I will give um because time time is

Running um so now this is from the individual to the group uh you have seen flocks of birds and this is an interesting um uh example because here you have starlings that flock really well you’ve always uh everyone has been a flock of Starlings I guess and this is

Not a Starling this is a fuckon uh and it’s midday so lunchtime and says oh there’s a flock of Starlings here maybe I could get a little Starling for lunch and it is the fastest bird in the world it goes at 300 km an hour a fuan against

One Sterling has the Starling has no chance he will get taken because this one is so much faster but here there in a group and this group is a complex systems and it’s a complex system and it has emerging Behavior and what you will see is that well the the Falcon is

Coming down but suddenly uh the group he was there and he couldn’t get one so he’s there again okay and he’s going to come down at high speed the group interacts with him and say no I will not g be able to get them um I’m not fast

Enough and because all animals are intelligent we’re going to see this in the next presentation this afternoon uh he went to University of hackens and he said oh I don’t have enough um kinetic energy I need to have more potential energy so it is the fuckon is there and

It is climbing climbing climbing he’s gaining potential energy for the f is in the room they know what it’s about and then when is coming from really high W he going to dive at full speed 300 km an hour and and he gets there and he couldn’t catch one

Okay and then well and these they are intelligent too and they say oh um there’s a around here it comes here again couldn’t get one and they say okay maybe it’s better to land and to it’s too dangerous to fly and so here’s an interaction you’ve seen between a

Complex system and an attacker all right now what is really nice is that the object of the science of complex systems is to actually model find the mathematical model behind it and um uh so here are birds okay and in a flock and we we well Craig reinold uh has uh

Found the equation between flux of birds in 198 5 they’re called boids and this is what he got so this is not Birds they not any other animals and this is a complete simulation of spheres inside the computer and they behave not exactly the same because the initial conditions

Are different okay but they behave in extremely similar way to what is there and here you because you’re an engineering school you’ve been asking me how to engineer complex systems and it’s very important there is something to learn about this it is the laws that

Have been uh used in order to model this um so here are the three laws of Craig Reynolds and this afternoon we’ll see the difference between laws and rules and if we have time so here you see that when you are a bird you only know about

Your direct um um uh neighbors you don’t know about hundred of thousands of birds that compose the whole group you just know about your five Neighbors in fact it is seven neighbors and when you’re outside you say oh I don’t want to be outside because maybe I can be caught by

A um an enemy I can be called if I a penguin so I want to be in the middle of my neighbors so this is cohesion I want to get I want have people around me and don’t feel secure if I’m if I’m alone on the border then the second thing is that

Of course um uh you don’t want to I’m going to get to this one there’s separation you don’t want to bump into each other because when you bump uh when you get hurt you cannot fly because the other one is too close or you cannot beat your wings so there is separation

So it means that when I am too close from that guy I will go away from it a little bit because I need to have my vital space between the different individuals and then um if I don’t want to bump into them constantly when I’m in

The Metro the tube uh if there is a a crowd of people going this way if I’m not going the same if I’m not aligning with the others I will bump into them into the others so there’s a third rule which is alignment roughly I will try to

Go the same speed same direction as the others and when you put these three rules together you’ve got a miracle that happens you’ve got the A Flock behavior that happens okay and it is homogeneous and it makes all these beautiful shapes now in the um engineering of complex

Systems what is very important here is that you’ve got a law that is Agonist mytic which goes towards the center of the neighbors and then in the same system you’ve got a law that does the opposite and this is an antagonistic laws and so you have two sets of laws

It’s called the ago antagonistic laws one law says I want to be in the middle of of my neighbors and the other one says oh yes but you’re going to bump so you don’t have you cannot be too close if there’s only this law without the

Second one all the birds will group and then you’ve got group and everything is static and they will fall because they can’t fly anymore and um it’s it’s going to to not evolve if you only have the separation one then you’ve got individuals that will go away one from

Each other but they will not want to be part of the group so here you cannot have a flock if you have only this one but if you have both at the same time you get a dynamic crowding system which means that you’ve got a bird that is

Going inside the group and then this guy he said oh he will he will be on the outside and say why why should why should it be me who is going to be eaten by the Falon so he will want to so the separation will put some individuals out

The individual out will say oh yes but I I want to go back in so and so you’ve got some kind of a dynamic movement okay which makes the whole flock to be constant constantly evolving and keeping its Dynamics so you must have into a complex system that you design you must

Have ago antagonistic rules you must think of rules that go towards One Direction and go towards the other direction to keep the Dynamics in societies uh the agonistic force is religion that creates groups okay and the antagonistic force that opposes religion is any Economist in the room it’s

Economy because with economy you want to go and see other people from other groups to sell what you have and then you are going to welcome someone else who because he knows how to make metal and you don’t know you you know how to make fabric so maybe you can exchange um

I don’t know um a piece of fabric against a piece of metal and by doing this you’re going outside of your group and uh by you take someone from the outside of your group also you can get marriages in in in between two different groups so in human societies it’s also

AO antagonistic laws uh one agonistic law is creating group it’s called religion and the angon antagonistic law it’s called um it’s called economy it is breaking groups okay okay so that is one of the things oh of course it applies with any kind of uh animals and so this is a kind

Of animal that you know well and here you’ve got a group of animals that are going to the left okay and there are some people in the middle they don’t want to move so the group is just going around us and the flocking behavior that

You have for Birds you have it for fish you have it for humans you have it for well any kind of animals all right okay and so you’ve got penguins that discover uh the notion of a circle but I don’t want to get into this so um then at the population level

You’ve got Evolution which is how you create new species uh I give this talk to biologist so and I know I’ve got a biologist here in the room so I’m going to ask him directly a question what happens when you you cross over a bird with a with

A yes what happens when you get this kind of crossover do you know what is the resulting animal as a biologist no well I have it it’s it’s there yes okay and of course it runs very fast and uh it pinches your but bottom really fast too so if if this

Thing runs after you you don’t want it to run after you and of course all the biologist they they laugh as because it doesn’t make any sense but they laugh a little less when I show them the next animal okay and you know what this animal

Is with sound so this gentle animal if he wants to dive this is called a platypus and when you have a look at what a platypus is uh maybe I’m going to remove the sound well a platypult is a much more of a mix

Than a bird and and a mammal um so first of all it has a beak like birds you don’t have many mammals that have beaks okay and then it has fur so it doesn’t have feathers and the fur it has is um um more belongs to mammals and then um

When it has uh Cubs it lays eggs so again it is you don’t have that many mammals that lay eggs so it typically it’s more birds that lay eggs but when the eggs hatch the hatchlings the babies they go back to the mother and they suck

The hairs and of there’s some kind of milk that comes out of it and mammal means uh animals producing milk females produce milk so they’ve been put into mammals because they didn’t know exactly what to do but it’s like cars you have options you know if you have an

Expensive car you can have a a car with GPS you have have a car with a nice uh high-fi radio system and this one has a couple more options um the males behind their uh left their their feet on their behind feet um they have um uh

They have venom here and if the dog is really bothering them they can actually uh uh Sting the dog and the dog will die so they are venomous do you know many venomous Birds venomous birds not so many venomous mammals there are some that are really restricted and then the

Beak is is a bit special too um the beak is in fact an electromagnetic sensor and when they are diving they close their eyes and they feel the electromagnetic environment around them and if there is a small shrimp behind that is uh moving it changes the magnetic environment and

It turns around it’s like it’s like fish you have eels you have sharks that have a u electromagnetic sensors like this and so this is much more complex than the example we saw just before and in evolution you keep having such incredibly uh incredible animals like

The op opo Pro there it’s a fish and um of big depth it’s these are not its eyes its eyes are up there and it is looking vertically through its skull and the skull is transparent so it’s something that is really bizarre and evolution keeps giving some most bizarre things like

This and uh uh you even have some guy who is talking to you that’s is very very bizarre and um so now we have an emerging Behavior coming from complex systems and how can you exploit it that is the big thing um so the big open

Problem uh it is how can we engineer complex systems and what is nice and this is also why I am originally a computer scientist I’m a computer scientist who is specializing in artificial Evolution and um it is perfect suited to massively parallel computer computer simulations why because we have many autonomous entities

And if we have a computer with thousands of different cores we can allot One Core to an entity and then how see how all the entities behave together and what is the emerging process coming out of it so the idea is that we can have an engineering uh the best engineering

Environment is to have a massively parallel computer um with thousands of cores right now you’ve got an Nvidia GPU card that has 10,000 CES and if you have an object-oriented language then you can create entities that are objects that are called intelligent objects uh that interact with others and then you can R

Run the whole thing and see what emerges you can have several kinds of objects to simulate Predator prey models you can have really really complex things and you run simulation and You observe the behavior of the whole and does the emergence or emergence correspond to what you want and if it

Doesn’t correspond to what you want what you can do remember to implement ago antagonistic laws you can use another complex system to evolve what you want to do and so you can use artificial Evolution and with artificial Evolution you get into a darwinian evolving complex system that uh will um uh evolve

Into doing the emergent property that you want so you can use complex systems to create other complex systems and it is some kind of uh loop that that you can put together here okay I think it is uh quarter to 11 so if I want to have a

So we can evolve virtual creatures we can evolve many many things so let me get to a conclusion uh so the aim of complex system science is to study the real world around around us in between infinitely small infinitely large by reconstructing models from observed data we want to find the mathematical model

How can we create a flock of birds how can we create this the objective is to understand the multiscale Dynamics on the frontier of chaos on the edge of chaos um uh when you have a liquid uh and um the liquid becomes a gas you’ve got a phase transition um when it is

Around 20° centigrade if you you add 5° or if you remove 5° water will still be water when you are at 99° Centigrade if you add one degree you could turn your water into a gas okay uh and so you will see as as you’re going

To go next to the edge of chaos what was stability could create a beercation could create a phase transition in the sense of ROM a catastrophe and then suddenly you can uh uh you’ve got a liquid that can become a gas or you can have a liquid that can become a solid

All right um and the idea is to predict in probability what could happen and not what will happen right now apart from mathematics mathematics are an different object they are exact the rest anything but mathematics is a complex system languages the evolution of uh of literature I’ve been president of a PhD

On compared literature literature is interacting influencing each other new literature is coming from other different kinds of literature it’s a complex system um and so the idea is to do this at the level of the planet the ecosphere climate economy Education Health and starting from observed data

Which is a participation it is possible to reconstructive to reconstruct predictive models the second P to simulate and try to prevent catastrophes in a personalized way okay we need to stop there thank you you and we have three minutes four questions no maybe until 11: we said well as

Please you have well now sure you’ve got the floor so you you manag the people so the session is now over for for questions please anyone Hi H thank you very much for the presentation let me say that this is a fascinating topic so um it’s important because uh in order to

Understand the behavior of all these chaotic system name it multi- universe so um no uh it’s important not to be in the chaos it’s important to be at the edge of chaos when you are a far from chaos nothing is interest interesting is happening in the middle of continents

You don’t have volcanoes why have people uh gone to uh um hulani and Pompei why have people been just where volcano are because volcanoes get back some interesting things in the earth so that okay it is on the edge of chaos that things are happening if you’re chaotic then it’s completely random there’s

Nothing to do with it it’s very important to stay at the edge of chaos not too far and here everything becomes really sensitive yes near chaos yes I understand so that’s make me think that it is important to to see an overview but at the same time a particular

Particular View and to see in a short period of time but at the same time in a large period of time to try to understand my question is uh it is possible to find a cycle uh cycle Behavior ER in all these chotic system no because otherwise they wouldn’t be

All unless you are in the stable environment where here you can have cycled but where it gets interesting it is at the edge of chaos and it is where you go from cyclic Behavior to chaos and you’ve got a an edge there it is in cellular automaton in computer science

You got type four automata that create life the the algorithm of life is there in cellular automatan it’s type four and so you’ve got things that are more complex meaning that they are not cyclic but they are not completely chaotic so they’re not completely random and here

At the edge of chaos there is something that is happening and it’s here that you should be if you want to evolve fast if you’re inside if you’re inside uh um um stable stable Ness nothing is happening you’re not going to change you’re not going to improve if you’re too far away

You’re in chaos everything will be destroyed and uh um you have to look at the sensitivity of the variables again at 99.7 de one10 of a degree is important for water to go into uh gas or not it’s to the tenth of a degree if you’re at

Around 50° 50.1 or 49.9 de will not make any change so you can detect how far away you are from the border of chaos by looking at the sensitivity of your variables when oh when the temperature becomes s sensitive at 100 of a degree then you’re going to then you are in in

Interesting place it’s it’s also worth for humans for universities they want to be at the edge if you’re far from The Edge the alter will overtake you if you’re over the edge ah the whole thing EXP load and that’s too bad you have to control where you are I don’t know if

This uh gave you some kind of answer to what you were thinking it’s okay thank you very much yeah another question um good morning ER thank you for your presentation um my question is about uh the complex system applied to education I’m a mathematician but I uh I work at

The ER teaching and the learning of math uh for our student in our faculty um do you think that the level of development in the education context is uh is just staring it’s just H or it’s it’s a high level of development and what what are the um

Er the angles that we have to look in that context okay um education is a complex system for the following reason if you are not a complex system it means that you have a student who is alone he’s not interacting with anyone else and maybe he’s got a book but if you are

Alone in front of a book and you don’t understand what’s in the book and you cannot ask anyone else for help then it’s going to be very difficult for you unless you’re a genius to understand what’s in the book and to improve so the idea is that in order to help people

Learn faster we created schools and we created universities and what is a university a university is a place where you have many different entities which are students who are given the opportunity to interact in between they can ask themselves questions but you also interact with teachers who are

Other kind of entities who that can give some um ideas give some knowledge and then there got exercises when you have a student who is um working on an exercise sometimes he fails and he can get educated and sometimes he wins and then

He can um go higher in a level and so up to now what humans have found that is working the best for promoting education as a complex system is to create schools and to create universities so you are in the best environment best complex environment that and the emerging

Behavior is that students learn faster when they are into an environment that is Rich with interactions this is how they can help it’s through communication so yes the answer of a complex educative system it is where you are right now which is University this is why we must

Develop universities and it is through interaction that people will learn faster and better and this is what we try we so we must try to um um uh favor interaction with uh I don’t know a cafeteria with a place where there’s a social interaction where students should

Should try to there should be a campus and in this campus you could have some uh evenings uh where they drink beer but this is also how research is done uh a lot of research come from people talking around a beer this is interaction and this is the emergence that you get so

You must try to maximize social interaction because this is what otherwise if you don’t have social interaction you’re alone in front of your book and well nothing emergent will happen so this is yeah thank you another question ah okay hello um you’re talking about multiscale Dynamics yes so

Um maybe uh to focus on what could happen but if the larger uh uh SC scales larger scales H maybe they could emerge to one one point also or not in your experiments or your data because if we we take first the the the the little scales maybe there is

Cowos and we don’t see the the the the relation in between but when we are talking about bigger scales maybe they uh yes so it is possible that they go towards point and you’ve got a very nice example which is uh human beings millions of years ago you had different

Species of humans you had uh I don’t know um uh homon nanes you had uh different uh um um um they were not human yet there were different species and that they interbred and now in 2000 we only have one species of human being beings okay we can all

Interbreed um the chrom manions and Ne neand cells don’t exist anymore and so the result of evolution which is an inherently complex system ended up in um removing all the different species of humans and now there’s only one but we are of course more evolved and we can do

Things that the other species couldn’t do so yes it’s possible that Evolution goes towards simplicity so so how you select the scale of what the first stage ah the first stage because if if you it’s atoms yeah if you take two little uh uh scales yes so this is why you must

Focus on the scale you’re interested in if you’re SC if you’re interested in education what is the entity of Education the entity at education level is the student it is the teacher it is the books it is exercises and you try to look at in in at at this scale what

You’re interested in if you’re interested in the body of a human then the entity you are interested in are all the organs and how they work together to see if uh the the human is in good health or is in bad Health then if you

Look at the scale of um I don’t know the liver um or does do the cells of the liver function correctly so at each point whenever you want to focus on a scale you have to identify what is the individual level and how so this is

Would be for one scale now to predict from one scale to another this is emergence then it’s more difficult to go from the second to the third and then because of course it’s it there are some ra ramifications everywhere and of course if you talking about the level of

Planet ecosphere yes is the biggest one so starting from the person or an atom is yes there are many different scales so there are around around 15 scales uh to go from the atom to the universe there have been around 15 that have been identified so you have 15 different

Scales it’s not it’s not that many but at every scale you’ve got something higher and of course the going from one scale to another one is uh some people have their whole research life on this I mean in the first part of the scales they diverge but then in the biggest

Part they can converge as you see for human beings at the scale of individual you had many different species millions of years ago and now they have converged and now you only have one species of humans so yes things can can happen and it’s not always towards complexification um okay there’s another

Story on this on complexifications and life of the second uh law of uh thermodynamics but that’s a something else yeah uh rodo maybe my question maybe for the next ah maybe for the next talk this afternoon but now for this one yeah in fact if we are thinking about

Evolution yes and the evolution of life and the possibility of Life yes I mean maybe they could be some uh I know there’s many books about that regard the possibility of life in other planets yes and there there’s contingencies yes so this contingency defines the evolution of the compx so

You’re getting getting back to what I was saying that I maybe I didn’t want to get back into this before lunch because it’s 11: no it’s the second law of Thermodynamics the second law of Thermodynamics is very interesting because it says that it should um maximize entropy you are a physicist you

Know about entropy and and but there’s something interesting is that it will maximize entropy at the fastest speed and you can measure the diffusion of entropy in water uh and if you heat your pasta every evening when you eat pasta okay you will put a um a pan of cold

Water onto a fire or onto a heating system and there you will get some uh um uh some cells R Benard cells that are um that will get the cold water which is warmer go up and then once it’s it’s War it goes up and then it pushes some cold

Water down and you’ve got the Dynamics of the complex system that is here through the ra bance now when you take the same pan of water but before you put some gel in it you jellify the water so by jellying the water you prevent the uh ra Bellar cell from happening when

You’re heating on the bottom the bottom layer of the gel does not go up because it is gelled all right and then what you would we see is that transmission of energy between the hot source and the cold Source will be much slower than if you have the convection

Cells right now if you translate this at the planet level there’s a very nice Theory which is that you’ve got the Sun that is a heat source and you’ve got planets that are cold sources and what physics wants to do physics want to transfer heat from the Sun to the cold source

And then you’ve got two choices either you’ve got a planet that is completely inert and that is um uh like mass for instance and then the sun heats mass at a very low rate because nothing can move on it nothing is dynamic or you’ve got the Earth and what happens in the earth

Is that at some point there has been life and what you get with life is that you get trees and trees what are they going to do they’re going to increase the surface of the Earth by producing leaves so as to attract more heat okay and then you’ve got animals and all the

Ecologists will tell you you shouldn’t eat beef because one kilo of beef beef uh is a resulting of more contains more energy than one kilo of vegetables and you see that somehow uh the life can be seen as the equivalent of raar cells that can increase the rate of diffusion of

Entropy in at the spatial scale and so um you can see the development of complexity as a way to increase the speed of the diffusion of entropy because many people said how is it that life becomes more and more complex if entropy means that everything should degrade it’s not true because

When you heat water you’ve got R Benard cells which are a structure which decrease entropy and you decrease entropy temporarily so that entropy can grow faster in the end so Planet so you can have if you’ve got the spark of Life at first then physics is interested in having a living planet

Because it will allow heat to be transferred faster between the hot sauce and the cold source and as humans we’re doing it really well with global Heat but that’s another story okay I think that we have to already maybe a very short uh um practical question because uh we despite

The the main ideas of complex systems uh as Engineers we always uh need to uh find parameters and stuff like that so AR okay finding the parameters is the next talk okay is the difference between optimization and machine learning ex exactly yeah my question was in that in

That uh so when when you have when you can uh use these black boxes to replay so you lose explainability but maybe you can have this next talk next talk we we’ll see a difference between optimization and machine learning yeah okay thank you very much Pier for this

Amazing talkk so we now have uh we need to take picture then we can eat some uh things outside and then we can move to the central campus of University Albert to thank you very much beer