Mindscape 242 | David Krakauer on Complexity, Agency, and Information
ฝัง
- เผยแพร่เมื่อ 31 พ.ค. 2024
- Patreon: / seanmcarroll
Blog post with audio player, show notes, and transcript: www.preposterousuniverse.com/...
Complexity scientists have been able to make an impressive amount of progress despite the fact that there is not universal agreement about what "complexity" actually is. We know it when we see it, perhaps, but there are a number of aspects to the phenomenon, and different researchers will naturally focus on their favorites. Today's guest, David Krakauer, is president of the Santa Fe Institute and a longtime researcher in complexity. He points the finger at the concept of agency. A ball rolling down a hill just mindlessly obeys equations of motion, but a complex system gathers information and uses it to adapt. We talk about what that means and how to think about the current state of complexity science.
David Krakauer received his D.Phil. in evolutionary biology from Oxford University. He is currently President and William H. Miller Professor of Complex Systems at the Santa Fe Institute. Previously he was at the University of Wisconsin, Madison, where he was the founding director of the Wisconsin Institute for Discovery and the Co-director of the Center for Complexity and Collective Computation. He was included in Wired magazine's list of "50 People Who Will Change the World."
Mindscape Podcast playlist: • Mindscape Podcast
Sean Carroll channel: / seancarroll
#podcast #ideas #science #philosophy #culture - วิทยาศาสตร์และเทคโนโลยี
Amazing episode. One of my all times favorites on this channel. I love how Sean sticks up for us and clarifies things as much as possible without "dumbing it down".
One of the best episodes to date. Fizzing with ideas. Thank goodness there's a transcript so I can learn more about the people and theories mentioned!
00:00:00 Introduction
00:04:02 Teleonomic matter and the origins of complexity science
00:06:16 What are the features of a complex system?
00:08:46 The milky way is not a complex system
00:10:27 Definition of complexity: Weaver 1948, Simon 1962, and Kolmogorov 1968
00:13:00 Balancing equilibrium and non-equilibrium
00:15:09 Adaptive systems
00:18:58 Methods in complex science: networks, hierarchies, and power laws
00:21:30 Is complex science paradigmatic?
00:26:00 Nested paradigms
00:28:09 Agency vs action principles
00:30:35 Natural selection and Maxwell's demon
00:36:24 How information comes to life?
00:38:56 How the second law leads to teleonomic matter?
00:43:21 Emergence: broken symmetries and effective theories
00:47:56 Downward causation
00:52:24 Information Gathering and Utilization Systems (IGUS)
00:56:22 The role of noise in complex systems
00:59:52 Information theory of individuality
01:07:08 Santa Fe Institute (SFI)
01:10:10 The future of physics
01:15:06 Non-linearity of science
01:18:23 Intelligence and cognition
01:22:14 Cognitive opacity
01:25:48 The future of complexity science
01:29:09 Scaling in academia
Thank you!
Anyone know if this has been done for other episodes? This is just fantastic 🎉🎉🎉
I have no idea what you're talking about, which is not uncommon, but your voice is soporific so I get great naps and gladly contribute through Patreon.
Not all are made equal
At least you understand the word soporific!
You are both honest and awesome 🦋✨
Great episode. Adaptation isn't action - it's intelligent interaction.
Oh, wow--this episode turned out being one of my favorites. Totally reframed complexity for me.
as a regular listener, this has been one of my favorite episodes. Thank you !!
I love David's voice, it's just jolly 😊
Consider when you have this string of information bearing degrees of freedom, it itself adds to the descriptive length of natural selection. It's part of its own environment. So theoretically there shouldn't be a real upper boundry?
Thank you Sean and David. Sharing the thoughts and paths of thinking that scientists are pursuing is motivational and encouraging to my own efforts.
This is the most interesting thing I've ever listened to in my life! I now have all the rest of Sean Carroll's podcasts to listen to and I'm so excited! Does anyone have advice on what order I should attempt these? Start with number 1?
My favourites are stuff on philosophy - he's got a great solo episode on morality where he builds up his antirealist/constructivist position and then applies it to show why Sam Harris is a dick. And how come it's probably OK to eat (some) animals. Robert Sapolosky is a favourite guest. But it just depends what you're into, the guests are pretty varied and Sean is an excellent interviewer.
"Mindscape 169 | C. Thi Nguyen on Games, Art, Values, and Agency" Is fantastic, highly recommend.
They're isolated, so let your interest be your guide.
The philosophy of science ones are my personal favorites! Maybe just watch them all, though--you'll get a sense for what type of episodes you like.
I enjoyed his episode with Michael Levin quite a lot
Fantastic conversation! Thank you for this!
This episode inspired me to mow the lawn and clean the gutters...I have no chance in the intellectual realm apparently. Lol.
That example about chemotaxis and the information gradient towards Whole Foods may not work with oranges and Whole Foods, but it does with the smell of french fries and fast food restaurants, coffee and cafés, cinnamon and icing and Cinnabon…
I liked Krakauer's interview with Cormac McCarthy
Great conversation! Thanks!
Wonderful discussion, and I have so many questions! More than a comments section could contain, I fear.
I definitely started off listening very skeptically of this topic as a whole, and now understand why it’s at least addressing important questions in a fresh and inclusive way.
I suppose the “t-shirt” would need to say something to the effect of:
“inside this t-shirt is an emergent effective agent of frozen accidents solving a vast range of problems carrying forth in time a modeled final boundary condition”. 🎉
It's tempting to wonder whether the kind of matter that can infer is related to entanglement of past and future matter.
Bloody fantastic podcast!
I just searched for the paper Krakauer mentioned “Life is Problem Solving Matter” and couldn’t find it. Anyone have a link?
Wow, Sean! What gives!? This was quite unexpected, but refreshing. You’ve stumbled upon one (large) prime number in a series of composites, pardon my metaphor… I really enjoyed the episode: a kind of revelation to me (not that I understood or was familiar with all that was mentioned and discussed, but I do like/align with the concept). One or two corrections (in opinion) are due - so, obviously, I simply must deliver them here… The utility of LLMs doesn’t lie in… Let me try another approach… I use chalk and blackboard (or abacus) in one way to assist me in expressing myself and in calculations with collaborators and in the presence of audience; I don’t “use” people the same way (it’s either impossible or it would be unethical). That’s the difference in tool use (say, simple exploitation vs. collaboration of sorts, although I know this will have a hard landing). Ways and means differ… There are different categories of benefits that float around this material world… And the generative AI is so potent when it comes to the ease of information fetching and summarization isn’t it? (BTW, Chomsky’s theory is so busted…) Second thing, about emergence… Either consciousness is emergent (in which case who’s to say the complexity and self-organization don’t play a crucial role - be it in carbon or in silicon) OR consciousness is fundamental (in our very own instance of consciousness-friendly universe, one of “many multiverses”). [Robert Laurence Kuhn said it first, except in all-PC voice.] I think tiles of epistemology (not to be confused with astrology), are beginning to align rather well if not yet perfectly. What can I say: We are inching closer to truth… (That is, our own, species-wide, not very cultivated nature may beat us to the finish line, but still…) In any case (of either mysterious fundamentality or epic emergence), traditional science will have to be revised… Funny thing… Even since my pre-teen days (in early 1980s), when I was asked what I will be doing when I grow up, I used to say, very confidently: “I’m going to be engaged with cybernetics!” Some of our well-informed family friends would then ask me if I knew what that was, and I would reply: “Sure; it has to do with computer systems…” In my mind that was the pinnacle of human intelligence and ingenuity. (Although, at the time, I had seen no such system, except in movies.) Well, The Universe reveals itself in lot of odd and probabilistic ways, but the essence is there - and was there - all the time.
The most intriguing thing I heard in this episode of your Mindscape podcast (one thing that eluded me until today, although it was in front of my eyes all the time) - and your guest brought it up: the defining feature of life is its internal representation (the operating memory) of the world at large - or something along this line of thought… So simple, yet so amazing…
PS: I just watched the other day: “Christopher Isham - Metaphysics vs. Materialism?”
th-cam.com/video/mQXe35xv5Ls/w-d-xo.html
“It’s almost impossible to get the real information.” -Tanner Stewart
How good was this comment, it has become harder to see the ongoing misinformation.
Thanks SciManDan and all the other debunkers.
I was always unconvinced by using terms like networks, power law in every discussion about complexity
I am suspicious of those that cannot explain things in a simple clear way and name drop right and left. To my mind he makes the simple purposely complicated . I had no problem reading Hidden Order by John J.Holland on complexity and chaos, but D. Krauker seemed to relish in obscurity so as to to hover over lesser beings.
And my admired Sean Carroll lets his guest, once again, get away with it.
the formation of stars and their burning are also forms of early life in the early universe. these processes are governed by the same laws of physics that finally produced biological forms of life and drive their evolution on a planetary scale, which have a highest level of complexity in the observable universe.
He mentioned a third paper after Simon about Algorithmic Complexity. I didn't hear the name of the author or paper, does anyone know what it was?
Found it in the Transcript: komogorov 68 which was algorithmic complexity
I see you🙃
Relative to the size of the universe there isn’t much Standard Model physicsy-stuff. That teensy-tiny bit of stuff is what you’re talking about here. So consider this…
•The universe came into existence in one instant with the exponential inflation of the quantum vacuum of empty space
•There’s a lot of energy expended in anihilation and a few remaining parts assemble themselves in a reductionist fashion: atoms, stars, black holes, galaxies, etc. No Maxwellian Demon necessary.
•The structure formed by this assemblage is now a factory that makes the quantum vacuum of empty space. Period.
•Scientific observation provides the evidence that there is no force in the universe strong enough to create another big bang from natural causes.
•Life and consciousness are emergent. Conscious life is the only option for creating a new inflationary event. Thus the universe has created the parts from which conscious life will emerge to create an event that will lead to the next generation of the quantum vacuum of Empty Space.
•That is the foundational evolutionary sequence which is your fundamental Action Principle.
Either I'm misunderstanding his point about Dawkins or I think he's fundamentally wrong. He said something like "Dawkins claimed that the gene was the only mechanism that can persist information" and then talked about culture and other ways of representing information. If he is implying that Dawkins said that culture could be reduced to genes that's absolutely wrong. The whole reason that Dawkins created the concept of a "meme" was because he realized that there needed to be some other kind of replicator to model things like ideas moving and evolving across culture over time. For one thing this type of change (e.g., the evolution of science since the renaissance) move at a much faster rate than the geological timescale that typically apply to genetic change. If he is saying there is some other form of representation besides genes that describes behavior such as eusocial insects. I think he is also clearly wrong. Eusocial insect behavior is understandable with traditional neo-darwinism. That's mostly what makes it "neo" darwinism is that before people like Hamilton, Price, and Trivers, we didn't have mathematical theories that could explain altruistic behavior that made sense in terms of genes, with new-darwinism we can and we don't need any special constructs for eusocial insects. The fact that they share a far higher percentage of their genes than most sexual organisms already explains why behaviors such as some organisms being born sterile or a drive to sacrifice one's life for the sake of the hive without reproducing can take hold and persist in a eusocial species but not in most species such as mammals.
infinity and beyond
The first thing he said that I disagree with is that literature and music may be some new kind of science independent of physics. Perhaps he’s being tongue in cheek but those types of things are subjective. I would argue passionately that Jimi Hendrix was better than the Monkees but I know some people would disagree with me and I can’t prove them wrong the way I can prove that someone who claims there is an algorithm to determine if any arbitrary Turing machine will halt is wrong.
Environmental conditions account for the chirality of the molecule or system. I’ll site, my sources here as well. This is an example of a system, in which how much of random chirality turns into homo chirality; th-cam.com/video/9fSh6uWrNMA/w-d-xo.html
Also, what you were describing was critical point theory that the system goes to a certain configuration then it’s a critical point, then goes to another configuration and hit another critical point and it stays there usually it before it hits another critical point, and the configurations is reconfigured
Noise, as an integral part of evolution, removes evolution's anthropic considerations
I think thatnothing can be known, hence I'm an idiot saying things becaue that's aparently not true
Mixing weird on this one or just me? -Sean's audio playing when it shouldn't 🤔
Please get David Deutsch 😊
Teleonomic homonculus does not have any answer. No one knows why our universe moves the way it does. The future is not contained within the present but some futures are more possible than others.
The thought that keeps popping up in my head is: What if the Quantum level is "not" the fundamental foundation upon which all else is built?
Perhaps Super-position is an indication of a deeper level which gives rise to what we term the Quantum Realm which is simply a transitional stage between our Spacetime Reality and the Sub-Quantum Level we have yet to discover. The perceived probabilistic nature of the Quantum Realm would not appear to be so weird as it would be the natural result of Boundary conditions existing between the Sub-Quantum foundation and our Spacetime Reality. One would not expect to find “certainty” in the ever evolving state of such a Quantum Realm.
Of course! That's what I was always thinking. Our world might just be arbitrarily stuck in a highly thermalized region of the sub-quantum level, while there might be all kinds of structures elsewhere. Although I doubt we could ever discover those. Imagine being stuck in the middle of the Pacific Ocean. You just couldn't possibly swim far enough to reach any island, so Okham makes you rule out the existence islands or continents, and says water and air are fundamental. I would expect reality to be fractals all the way through, an automaton expressing pi digit by digit for eternity, or something like that, and even that wouldn't be fundamental, because how could it be?
He rattles off names and history much better than Eric Weinstein
Id like to ask this guy if he considers the universe a complex system. I sispect, his answer would be no. However im sure, he thinks himself complex.
I thought he addressed this during the interview? What I understood was that he didn't consider the physical universe a complex system because it's not adaptive? Maybe I misunderstood.
I got very angry when Sean asked about goals and teleonomic matter and you brushed it aside. Sigh. That's the Crux of his belief system.
I'm very suspicious about those who promote complexity. Tens to boil down to well articulated nonsense
This episode went a long way over your head
Milky way isn't complex??????
Sorry but i'm out....
Big doesn't mean complex. A bacteria is magnitudes more complex than the milky way once u know anything about cell biology
@@HonkletonDonkleton a bacteria itself is a subsistem of the milky way...
I suspect he'd say the universe is not a complex system
@@diegoalejandrosanchezherre4788 You have made a category error
@@AdamTait-hy2qh well, the entropy of the milky way is way bigger than all of systems presents here in earth (if You Want to categorize earth like an aislated System, thing that is think is wrong...) so the lack of information about the all milky way is bigger than any object of study here in earth, un consecuence is more "complex" acording to the definition of entropy, also taking on account the difinition of "quamtum complexity".
Wee need way more quamtum Gates to reach a particular state like the whole milky way, than any subject of study here on earth....
So the free statement that milky way isn't complex sounds like a joke for me...
Thos guy never really explained what hes studying. I don't think he knows what he is talking about. Just boring and trivial
Maybe you don't understand it.....maybe it is too complex for you....?😆🤣
First!