Тёмный

Biology as Information Dynamics - John Baez 

Stanford Complexity Group
Подписаться 17 тыс.
Просмотров 25 тыс.
50% 1

If biology is the study of self-replicating entities, and we want to
understand the role of information, it makes sense to see how information theory is connected to the 'replicator equation' - a simple model of population dynamics for self-replicating entities. The relevant concept of information turns out to be the information of one probability distribution relative to another, also known as the Kullback-Liebler divergence. Using this we can get a new outlook on free energy, see evolution as a learning process, and give a clearer, more general formulation of Fisher's fundamental theorem of natural selection.

Опубликовано:

 

28 сен 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 34   
@jonabirdd
@jonabirdd 5 лет назад
The slides are full of basic errors. 8:20: negative sign should be positive 10:26: k should be below the division sign. His use of notation I(q, p) is confusing because this is normally reserved for mutual information. (e.g. I(X, Y)) He also keeps changing which distribution he varies. Is it the first or second? I was slightly more enlightened but not satisfied. This article is far more insightful: www.nature.com/articles/s41598-019-45619-9.pdf
@withanametocome
@withanametocome 4 года назад
Please upvote Chuang's message so that it appear at the top of the comments and no one else get confused
@duartesilva7907
@duartesilva7907 4 года назад
@@withanametocome I did that!
@forwardescape2955
@forwardescape2955 3 года назад
I don't know if he's right so I'll just see where the chips lay
@unknownchannel3141
@unknownchannel3141 3 года назад
The real question is, how do you know this stuff?
@forwardescape2955
@forwardescape2955 3 года назад
No the real question is why do you know this?
@zacharyburell1747
@zacharyburell1747 7 лет назад
@24:44 the Dude Abides
@hyperbolaisagraph
@hyperbolaisagraph 6 лет назад
The editorial "we," man!
@syremusic_
@syremusic_ 4 года назад
Stupid question: when he refers to log(2), does it mean log10(2)? Or ln(2)? Would love an answer. Thank you!
@Suav58
@Suav58 4 года назад
in information theory one uses logarithm base 2. Generally log_b(b)=1. If you want to benefit from lectures like this, then do your precalculus, calculus (real and complex, of course), fourier analysis and fucntional analysis, maybe process control. Do some of these, and then think about limitations of applicability, which will bring you to computation theory and finitistic mathematics. You are probably a young person, so you have all the time to do it. Don't get put off. The ideas, that I mentioned didn't come to people in a fortnight. The great, which discovered them (seldom as mere individuals; almost always in social context) were themselves walking in the mist. Some of them did admit it.
@syremusic_
@syremusic_ 4 года назад
Slawomir P Wojcik Thank you very much for taking the time to respond. I have noted down all the concepts you mentioned, and will keep track of learning them individually as I go through my classes. I’ve got a long way to go since my fascination with mathematics grows each time I learn a new concept. Thank you again.
@Suav58
@Suav58 4 года назад
@@syremusic_ Hi, I've got to apologise: logarithm base b of b is one (log_b(b)=1 because b^1=b - that is b to the power of one equals b - that's and logarithm is an inverse function of an exponential. Sorry. Stefan Banach once said that mathematics is about seeing parallels and good mathematics is about seeing parallels between parallels (or something very similar). It means, that you should never take the subjects separately. There is a whole new subject of mathematics called Category Theory, which deals with the structures common to all mathematics (a rather esoteric book on it is Emily Riehl's Category Theory in Context)
@Suav58
@Suav58 3 года назад
@rwalser Depending what you mean by "understand". Maybe, just maybe, there is some way forward in "understanding" this lecture and correspondence between complex and p-adic analysis is vital to understanding of the developments? After all, the lecturer only points out to some toy models, while attention of a good (better?) part of scientific community is on evolution of complex systems and mentioned above are, to my knowledge, best tools we have.
@ManyHeavens42
@ManyHeavens42 3 года назад
Frequency times time times energy. 12 dimensions. Are you excited yet,this is why Quantum computing will save the universe! We have new eyes and new ears Now, We are God's now ! almost. Soon as we get over the fear of dying ! By Cloning and start Living.
@Tadesan
@Tadesan 6 лет назад
I love this man! You look great John! Thank you for bringing me such Joy!
@Joao-uj9km
@Joao-uj9km 3 года назад
It's interesting the red queen thing can also be applied to velocity in Minkowski space, since not moving in space means moving at c (as fast as you can) in time.
@jamesglorbin9976
@jamesglorbin9976 2 года назад
5:21: The second term in the computation of I(q, p) should be 0 log (0 / (1/2)), not 0 log(1 / (1/2)). Makes no difference to the computation (since the term is multiplied by 0), but for anyone new to this formalism trying to work through the equations themselves, this might trip them up.
@Johnnyredtail
@Johnnyredtail 4 года назад
This presentation put punctuated equilibrium in perspective for me. Thank you John. csmgeo.csm.jmu.edu/geollab/complexevolutionarysystems/Documents/Ramps2.pdf
@ManyHeavens42
@ManyHeavens42 3 года назад
Answer: What is Neutral! Eureka.
@AishaFaraj-bd3wl
@AishaFaraj-bd3wl Год назад
Hi John Baez Could you please give more information about negative temperature in finite Hilbert space and its probability and thermal distribution ?
@mohammadsultanalam8334
@mohammadsultanalam8334 4 года назад
Can anyone tell me how to compute fitness $Fi$ and the distribution $q$ (constant value)? Thanks in advance.
@shanemackinnon3575
@shanemackinnon3575 4 года назад
"this lecture will teach you everything you need to know about bitcoin"
@Luck_x_Luck
@Luck_x_Luck 2 года назад
answering before watching: nothing, amirite?
@AA-gl1dr
@AA-gl1dr 3 года назад
Amazing
@ManyHeavens42
@ManyHeavens42 3 года назад
Everything has a purpose And we're the smartest Then why haven't we learned to create ourselves, Will GOD Kill Us First,Is it our Right to Live ,That is the Question: Well do you Punk Hahaha
@danbaurceanu129
@danbaurceanu129 6 лет назад
Lots of equations, but are they REALLY related to actual biology?
@PeterBaumgart1a
@PeterBaumgart1a 5 лет назад
Yes
@Suav58
@Suav58 4 года назад
Don't tell, don't ask policy here. If you want to know, (and you do, because much more sophisticated versions of them are applied in, say, virology) then there is no other way, but to learn solving differential equations. Newton and Leibniz might be the first who wrote differential equations formally. It is very easy to write a differential equation (especially a homogenous, liner first order equation). It is a much harder business to solve one. On the other hand, you might be confinig yourself to watching a slightly upgraded version of cat videos. As for the one at hand you were given a suggestion, that (the space of solutions of) the equation can be moved to a Riemanian manifold, treated as a many sheet surface, which in itself opens a huge richess of possibilities. This is something for hard working people. Again, when applied to the video lecture, a perturbation, which gives the lizards higher premium for aggression (environment rich in food and warm enough to give them premium over theromstasis keeping creatures) might result in the blue ones evolving into dinosaurs again. Interestingly, the reverse is, due to difficulty of limiting ones size, nay to impossible. Latest changes in our political system can be modeled as a direct result in such perturbation of replicator equation. So, yes, for a man with a manifold everything is a dynamical system. This is a best tool set we have, so, probably worth knowing.
@giuliocasa1304
@giuliocasa1304 5 лет назад
I'm not sure Baez is mathematician. He is rather a marketer of some subfields of math. But of course he is not titled to speak about life and biology. That's the crucial issue with this guy: he is unreasonably proud of himself that he can't listen to and understand and ultimately care about the opinions of the others.
@philly1729
@philly1729 5 лет назад
Regardless of whatever he is talking about in this video, John Baez indeed qualifies as a mathematician. By "mathematician", I mean any reasonable definition of "mathematician" that a professional research-active mathematician would agree with. For example, he proves theorems in his non-expository papers, most of which are published in quality, peer reviewed mathematics journals, and many of these are highly cited. (Source: I'm a professional mathematician with access to MathSciNet, the AMS citation database.)
@giuliocasa1304
@giuliocasa1304 5 лет назад
@@philly1729 So.. yep, he is a professional mathematician like you: he comes from an ivory tower. But thanks to you for not publishing a video about biology, for not sharing wrong ideas about life and, last but not least, for not twitting that "computing Kan extensions is something that's actually very practical for migrating databases" (and many other similar, silly things). Also Trump qualifies as President for the number of American voters and he is highly cited by the media, but is not someone I would appreciate as a politician, regardless of whatever they are saying about him. Take care
@deepbayes6808
@deepbayes6808 4 года назад
Can you indicate what's the main issue with what he is saying? Is it that replicator equation is not representative in biology or what exactly wrong with his statements?
Далее
22. Emergence and Complexity
1:42:30
Просмотров 839 тыс.
ДЕНЬ УЧИТЕЛЯ В ШКОЛЕ
01:00
Просмотров 1,2 млн
Wolfram Physics Project Launch
3:50:19
Просмотров 1,8 млн
21. Chaos and Reductionism
1:37:33
Просмотров 1,7 млн