Multiway Systems as Models to Understand Mind and Universe - a Conversation with Stephen Wolfram
HTML-код
- Опубликовано: 28 май 2024
- Our earliest models of reality were expressed as static structures and geometry, until mathematicians of the 16th century came up with differential algebra, a framework which allowed us to capture aspects of the world as a dynamical system. The 20th century introduced the concept of computation, and we began to model the world through state transitions. Stephen Wolfram suggests that we may be about to enter a new paradigm: multicomputation. At the core of multicomputation is the non-deterministic Turing machine, one of the more arcane ideas of 20th century computer science. Unlike a deterministic Turing machine, it does not just transition from one state to the next, but to all possible states simultaneously, resulting in structures that emerge over the branching and merging of causal paths.
Stephen Wolfram studies the resulting multiway systems as a model for foundational physics. Multiway systems can also be used as an abstraction to understand biological and social processes, economic dynamics, and model-building itself.
In this conversation, we want to explore whether mental processes can be understood as multiway systems, and what the multicomputational perspective might imply for memory, perception, decision making and consciousness.
About the Guest: Stephen Wolfram is one of the most interesting and least boring thinkers of our time, well known for his unique contributions to computer science, theoretical physics and the philosophy of computation. Among other things, Stephen is the creator of the Wolfram Language (also known as Mathematica), the knowledge engine Wolfram|Alpha, the author of the books A New Kind of Science and A Project to Find the Fundamental Theory of Physics, and the founder and CEO of Wolfram Research.
We anticipate that this will be an intellectually fascinating discussion; please consider reading some of the following articles ahead of time:
The Concept of the Ruliad: writings.stephenwolfram.com/2...
What is Consciousness?: writings.stephenwolfram.com/2...
Exploring Rulial Space: www.wolframphysics.org/bullet...
Multicomputation: writings.stephenwolfram.com/2...
Multicomputation with Numbers: The Case of Simple Multiway Systems: www.wolframphysics.org/bullet...
Chat log for the recording: bit.ly/3xqe8tB 
Наука
Two of the greatest minds of our era!
Please make it a regular thing. Wolfram & Bach discussing bleeding edge theoretical science is priceless.
Once a week!
@@teemukupiainen3684daily 😂
this was brilliant - especially the fact that it was so interactive between both these people. I loved that format - hope they speak again.
I've been hearing Joscha reference Wolfram's ideas for a while. Happy they finally got to talk. I hope that they can talk again.
I'm so happy for Joscha that this happened
I love this video! I'm so glad that Joscha & Wolfram had a recorded dialogue!
For almost 2 hours I was in a state of euphoria! Thanks Joscha!
Same 😌💭☘️
Same.
I HAVE BEEN WAITING FOR THIS!!! I literally feel like I've witnessed a miracle
Like Mozart and Beethoven meeting and playing duos together. Than you so much. 🎻🎻🎻
This is so incredibly great. I really hope Joscha and Stephen talk again like this in the future.
We need a discussion between Wolfram, Levin, and Friston
My two favorite geniuses in conversation
😌💭
32:29 lol joschas eyes when him and Stephen were talking over each other made me laugh
I’ve used multi-way graphs and multi-computational processes to model a bunch of systems over the past year in studying Wolfram, and they are very intuitive and IMO they are the best way to model things.
How consciousness enters the story, barring a lot of details, is that consciousness is universal, but experience is localized. You can think of it like this…where you play a cellular automata like Game of Life. You see some cute structures appear…and these structures are experiencing their computational world, from their own localized viewpoint…but consciousness is bounded to the entire computer itself… in other words you as an automaton in this computation is that you are just observing yourself.
Me and you are talking… you and me are both automatons in this computer so you and me are just the same (computer) where different local parts of it (automatons) are probing each other.
You can go a bit further when you consider the deeper consequences of this theory…that like how josh mentions in the beginning, that the universe is potentially running all possible rules and will create all possible physics, and thus all things can exist. In Wolframs theory this is indeed the case, and if you understand this deeply you realize that there is configurations in which I experience your stream of experience , you experience mine…and we experienced everyone that’s ever existed…and everything that will ever exist we will eventually experience the world from all those view points…so in this sense you’ve lived my life, and I’ve lived yours, at some point in our past or at some point in the future. The caveat is that, we can not in a way “know” that you’ve lived my life or I’ve lived yours…because our experience is attached to our local “memories”
Interesting thoughts… closely considering…
What pre-requisite math & science knowledge would I need to understand the subject?
@@XxfishpastexX This is a bit of a tough question, because to understand the physics model by Wolfram, you don't really need any math or really any science. However, knowing all areas of science and math, helps you see why Wolfram's model is true.
The most basic thing to look into is Complexity Theory 101, which you can search for some video's here on RUclips. Complexity Theory uses a combination of Network Theory and Set Theory to make formal arguments about things.
Complexity Theory is a very broad theory that extends into many areas, which includes Chaos and Nonlinear dynamical systems, to Biology and Evolution, to Thermodynamics and Entropy...you'll quickly realize that complexity theory is a sort of connection between all these sciences but for 40/50 years it has gone without a formalize them into a single theory. Wolfram's model is therefor a formalization and complete extension of Complexity Theory.
You can also try to approach the subject from a historical standpoint. Science began with Newton and his attempts to quantify gravity using Newtonian Mechanics. In order to do this he created Calculus, a kind of math that takes discrete chunks of things, and granulizes the system toward an infinite limit to get a continuous solution to his problems. After having massive success with this model, the issues begin to arise with the problem of chaos...which is embodied by "the three body problem" a problem that is basically much harder to solve using simple equations.
From Newtonian Mechanics followed thermodynamics, which is the a kind of formalization of how discrete systems can be described by a global statistical answer. This is where the idea of temperature comes from...particles bouncing around and the average amount of movement is what we think of as heat. The issues that crop up in thermodynamics at first appear to be different than Newtonian Mechanics...but subtly they are part of the same problem...which is the role of observation of the system.
After Thermodynamics came General Relativity, where space and time were unified under a single mechanism which we think of as curvature of space...again like the previous problems...the issue of the observer and how all observers are relative to one another crops up again.
And Lastly is the era of Quantum Mechanics...same story yet again...the problem of measurement and how the observer plays a role in what we can do to these systems.
This issue that pervades all the subjects is the role of observers and how they take part in measuring and quantifying systems. So it is in Wolfram's model of Physics, that these relative effects are caused by the equivalence between all systems. This equivalence is the same kind that you see in say space-time relativity. Space and time stretch for local observers in order to keep the speed of light a constant. In that token, The principle of computational equivalence is that all computations are equivalent in their sophistication, and thus, observers that are observing these systems, are bounded by their capacity to compute them, and as a result you get the same kind of "stretching and warping" for how observers can measure these systems.
For example, you have some kind of complex computation taking place. Because you are computationally bounded, you can't out-compute the system because you and it are at the same level of computational sophistication. Thus the only thing that can predict the behavior of the system is the system itself. And your model of the system is not going to be able to accurately predict it. Thus the thing that is "stretching and warping" here in our relativity analogy is our ability to measure systems and predict them.
So I apologize for the very long post, but you can see why it's such a vast subject. I've studied Complexity Theory and all these other sciences for around 8 years, so I was able to jump into it and understand it quickly.
TLDR: Just to repeat the important bits of information, Start by learning Complexity Theory. It will give you a well rounded understanding for why Wolfram's model is the way that it is...alongside Complexity Theory, if you want to formalize the concepts you will spill into Network Theory and Set Theory.
As you explore Complexity Theory you will quickly find yourself looking into Chaos and Nonlinear Dynamic Systems, Thermodynamics, Biology and Evolution, and Computational Complexity...and of course Quantum Mechanics And General Relativity.
After analyzing the core problems in each one of those theories, this is when you will truly understand the importance of Wolfram's Theory which unifies all of them and addresses their issues.
@@NightmareCourtPictures thankfully, i’m familiar with most of the foundational math, except for tensor calculus. For studying Wolfram’s model do you use wolfram’s language/mathematica? python? How do you practice?
@@XxfishpastexX Great, I'm glad because this is gonna be a lot easier to explain then heh.
Wolfram's model is built on Complexity Theory, which uses Network graphs to model things (Network Theory and Graph Theory). So you have a set of things (called nodes or elements), they are connected to each other in some way (relations), and the way they are connected creates a network.
Wolfram's model uses a version of graphs called "hypergraphs" in which instead of using lines to create connections between nodes, you use "hyperedges" which connects sets of nodes with arbitrary number of edges...and you draw this as just circling the nodes like you would see a membrane around an amoeba.
And so this modeling can simply be done with a pen and paper. Whatever the system is that you want to model, you just need those tools : some nodes and some connection between them...but you do need one last thing, which is a rule that governs the behavior for how the relationships between these nodes changes, then you can compute in time steps what that system does and how it evolves.
So basically setting up the system of nodes is your kinematics, and then how you set up the connections at each time interval based on a rule, is the dynamics of your model.
Just an example. Say I have 5 nodes where they can take either a green color or a red color. The system follows a simple rule:
"A node checks it's nearest neighboring node. If it is green it turns red, if it is red it turns green. "
Not the best rule to pick...but just bare with the example. If you are using a regular network graph, draw a connection to the nearest neighbor, and color the cells according to the rule. Since the rule is "check it's nearest neighbor" all the cells are going to be colored by the end of T=1. Also since it's nearest neighbor, your gonna end up with the cells periodically switching between the two colors since the position of the closest neighbor never changes (lame rule sorry). But if your rule is more exciting, the rule might take on more complex and interesting behavior at each time step...then you analyze the systems behavior.
So that's the basics, and if you are using hypergraphs, you're going to have an easier time. Reason being is that hypergraphs have the curious but convenient property that hyperedges allow you to permute different outcomes of an arbitrary number of connections between the nodes...or rather nodes can be in a superposition of different possible connections, and the hyperedge expresses visually, this notion of superposition. It's this notion of superposition that allows for a more accurate representation of a real system rather than an idealized system in which definite things occur.
So for example, say you have 3 binary units, which can take on either a 0 or a 1. These 3 binary units are in a superposition of 8 possible states : 000, 001, 010, 100, 110, 011, 101, 111.
Modeling that as a hypergraph, you'd draw 3 nodes, and then you circle the three nodes like an amoeba. These three nodes are in a superposition of 8 possible states or 8 possible relationships they might have with each other...where one relationship might be 0-0-0, and another might be 1-1-0 and so on...
If you want to go a step further with your modeling, you can model the evolution of that hypergraph as a multi-way system which is just a graphical representation of the evolution of a hypergraph system, where states are in superposition.
So you know...say you have your 3 binary units again, and out of those 8 possible states they take on a definite state of 000 at the first time interval according to some rule. On the second time interval according to that rule, it becomes 011...then it becomes 110...then it becomes 000 again. You can graph out using a multiway graph, the 8 permutations, and then draw the evolutionary path of that system :
000->011->110->000->011->110->000... and so on.
The system since its in a superposition of 8 possible states, will have all those states there in the graph, but the rule that you applied dictates which evolution/path will actually occur. Using a different rule, you could have had a different evolution/path but the graph still contains all the states. This comes in handy...and is actually integral when you are analyzing a system where there is more than one rule occurring at each time interval!
TLDR: All the tools you need to get to model these things from Wolfram are in network theory, graph theory, a pen and a paper...hopefully this comment cuts down on the amount of searching you need to do to get to the meat of your journey. It helps to also watch Wolfram's lectures on his RUclips channel, his series on his book New Kind of Science, and reading his journals. Lastly, you can program the things you need into a computer program to run systems for you, if you have that knowledge. So Mathematica and Python are both things that can assist with your study, and probably make it easier too.
Stephen Wolfram is pure brilliance
Joscha is an amazing mind! Always great when he shares his knowledge with us.
It was a great back and forth. This was extremely helpful in understanding the Wolfram model. Most academics don't question Wolfram on his model. Instead they want to see how they can conflate contemporary physics into his model. Not the actual nature of his model!
@@pascaljosiah6866 they don't question him because they don't (and likely never will) see anything to gain from it. What's the point? It doesn't deepen or add anything to our understanding of ourselves or reality.
@@TheFrygar I mean, there is always the possiblity to gain new ideas from taking a different perspective. Maybe not devoting your life to understanding what Wolfram does, but inspiration all the way.
@@Dante3085 The ideas are already fleshed out...we've gotten as specific as we possibly can in the direction of low level physics. Creating yet another fundamental theory of how the tiniest things work which is consistent with all the other theories that have been around (and good enough for practical use) for half a century is not going to change anything about our actual understanding of reality. Saying "everything that is possible, exists" just doesn't do anything for us...as much "fun" as it may be to entertain the idea for 2 hours.
@@TheFrygar Many of the ideas and potential applications weren't fleshed out, not sure what you're talking about. Potential for proving that both relativity and quantum mechanics are two sides of the same coin within one unified theory is a Nobel prize material on it's own.
Great conversation! You can see how much respect Stephen has for Joscha.
This is one of the most interesting things that happened for so long in the intellectual internet space!! This conversation (Bach-Wolfram mainly)
was one of my wishes for long time ! THANK YOU ! :) :) :) :) :)
tbh: the screen is a waste of space and attention - I just closed my eyes and listened. This works better as an audio-podcast.
These two talking is all I ever wanted (so to speak). If I could add one year of lifetime to Joscha’s and Stephen’s lifetime’s by giving ten of mine, I’d happily die right now on the spot.
What a treat!
The fact that any Turing machine can be implemented within any same Turing machine can be the reason why we are conscious. Our brains are complex machines which have evolved within a complex computational universe, therefore the question of our observational limits are the question of the limits of the properties between any type of Turing machine implemented inside another such Turing machine.
would be good if only there two videos were showing and not all the rest ;-)
I loved it and look forward to these two talking again
I am so happy that the universal umbrella of my
being allows this beauty of flowers to sprout.
Should put Joscha’s name in the title
To get more attention on this,
Ty great upload
This is amazing, thank you for sharing!!
What a confluence 😌💭 thank you for setting this conversation up!
This makes me happy and comfortable
Great convo guys 🔥👍
1:34:00 beautifully stated.
About time these 2 came together .... A+
“How does a branching brain perceive a branching universe?”
I like the notion that the universe is an ultra-superconductor. Where gravity is a just a rule to help keep things cool.
Did Wolfram just solve the Fermi paradox?
I think Joscha gave it his best but then again Wolfram is about Wolfram and not about the ground-truth
You can generalize equivalence as the distance metric.
I need a definition of number based on factorials, please. It should be dimension without rank or order.
There's a Continuum end a discreet thing picture must be because that is the two different sides of the duality
It is the information that creates locations in a roundabout fashion at least because it is information manifesting itself as Consciousness creating moments in time that we know of is the present or now these locations in the temporal dimension of our universe that gives rise to locations in the spatial dimensions which give rights to all manner of other growth of infinity like Multiplicity add magnitude. All Infinities melt into all others
At the 29:30 mark, I didn't follow why Wolfram asserted that if you move a physical object, it's made of different 'atoms' of spacetime. He seems to be asserting the spacetime nodes are in fixed positions, rather than mobile nodes which collectively form spacetime?
In order to have a notion of movement you need to have a background to move in. So yes the nodes in the graph representing space don’t move they just transform according to a rule. It happens that certain structures persist through the transformation on space. If you wanna a simplified version of it you can see 2d cellular automata or conway game of life gliders.
@@Khawalidmi Hmm, if I move a graph in a computer, I may just reassign pointers. But you could be right: moving the object requires motion, and that happens by applying the rules that constitute physics, and the movement does take time, and there may be other forces at work, so thinking about it more I think he chose his words very carefully and meant exactly what he said.
Thanks!
I interpreted it as nodes in a graph. Each node representing an atomic unit of space (irrespective of place and time). And each connection of this node to another node as “atomic proximity”
And i interpreted Stephens notion of (atomic) “time” as a state transition of one state of this universal hyper-graph, to the next
I was always too scared to be laughed at to post this naive intuition: the air becomes my hand when I move it around?
joscha dressed up because he has a crush on wolfram
Pompous i might but I've been thinking about this non-deterministic turing machine that yaska is talking about for a few years now and I call it the singularity or the point of paradox point where the present moment of now is occurring or the point where all opposites are I brought back together where nothingness and Infinity Touch. All of reality is a pattern
The Wolfram model is close to the concept of emptiness in Buddhism if the IDs in the Wolfram model are just unique identifiers for the connections in the graph. Then one can say that reality is just relations connected into a graph (network/web).
❤️❤️❤️
Remember perceiving is this the opposite polarity of conceiving
How do you know that the many worlds interpretation of quantum mechanics is true ?
So, Entropy is non deterministic TM, is that right? I love it (:
Yes, I will be thy priest, and build a fane
In some untrodden region of my mind,
Where branched thoughts, new grown with pleasant pain,
Instead of pines shall murmur in the wind:
...
And in the midst of this wide quietness
A rosy sanctuary will I dress
With the wreath'd trellis of a working brain,
With buds, and bells, and stars without a name,
With all the gardener Fancy e'er could feign,
Who breeding flowers, will never breed the same
- Ode to Psyche, John Keats
is this a most important talk of our age?
When you speak of a Continuum you have to speak of infinity because that's what it is or at least potentially so because the universe will never stop expanding contrary to popular belief and one of these days the blank spaces will be as big as this universe is and they will all contain their own universities within them that have their own spaces destined to become the universe
I got this .
1:19:00
That's why everyone thinks of Consciousness as movement skill that realm of information that is not exactly complete Stillness and that is what mind is and that's what's infinite
i dint have a fucking clue what they were talking about
The reason anything existed at all is because it has to because information. What would describe what nothingness is except information? Isn't that something? So it is the default. But what is information and where did the information come from? Is it information a language? What could speak such a language?
How bout this: If you were given a PC and you opened it up, where would you find Windows 98? That is the problem we find when we introspectively think about our own brains and what our thoughts are about what our brain thinks. Where is the operating system? ME! I am the operating system of my physical body/mind! I don't exist unless I think I exist. The rest of the time, I'm just having fun.
Joscha is seriously on another level. Stephen wolfram is a genius, but the clear, concise, logical yet beautiful things joscha says makes everyone else seem slow.
2
32:16 ? Who?
The open AI image generating model, “DALL E 2”
@@ac4740 if anyone would like assistance knowing how to talk to DALL E 2 I would love to help.
@@ac4740 DALL E 2 is an image generator. Why don't they just ask the AI to describe it. Wouldn't that be much more useful?
@@user-vi7jn5ph9b neither a text nor image generating modern AI will actually be able to tell us anything new about the internal structure of an electron. might as well generate a cool picture.
@@ac4740 This is not true. And Joscha agrees with me or he would not of brought it up.
How does a seed become a tree? Inner and outer, that's how. Inner DNA, outer reacting with environment. Wolfram has planted the computational seed, but we must admit those with lesser thought capability will abuse his naive computational breakdown.
I think Wolfram's model is very promising, but incomplete. It respects the indeterminacy of QM too much. I don't see how it could be that if one were to observe the universe from outside of it, one would see multiple different threads of history. That seems like some weird multiverse interpretation which fails to properly address the all too real wave-function collapse.
Nicely expressed! Agreed!
I don't think it does respect the indeterminacy of QM - it is a lower level description that expresses the possible different outcomes as different possible sets of transformations of the hypergraph, which in turn occupy a region of rulialspace. It is still deterministic, but it describes how all possible sets of transformations exist in different branches of branchial space, which corresponds to a region of rulialspace - so there is a higher level graph that is the graph of all hypergraphs (and their corresponding branchial spaces). We happen to exist in a universe that corresponds to one of these branches in one region of the ruliad. I think the discussion about encountering alien civilisations that might be operant out of a different region of the ruliad is flattening out the ruliad and supposing that we might be able to transport to (and define translation mappings between) other regions of the ruliad.
There is no indeterminism, just different branches of branchial space in different regions of the ruliad.
"Frobulous leaf" 🤣
Everything is literally a measuring device
My guess? Absolute nothingness is impossible. It never occurred. So, the question, why there is something rather than nothing, makes a flawed assumption about nothingness. This makes the question a poorly formed question.
Yes nothing cannot exist by definition, if it existed it wouldn't be nothing, but something-nothing. Even the word nothing does not refer to nothing, nothing is a concept that conceptualizes no-thing, the very conceptualization of nothing is not real nothing, because nothing cannot exist, but this nothing that it is claimed not to exist is not the "real" nothing, which does not exist. The paradox is the ability to talk about something which doesnt exist, how is it possible to talk about nothing?
I don't like the guy who keeps interrupting
Wolves travel in packs - Rams butt heads - Bach was an excellent composer. Can we not agree that machines are going to take over?
Wolfram vs Bach on the Continuum. Bach argues wrongly against it saying that we observe at any vertex in Wolfram's Ruliad is only a collection of discrete things. (This is bad logic akin to Zeno's Paradox that the trajectory covered by an arrow will never get you there since the distance of 1 is composed of the parts 1/2 + 1/4 + 1/8..(distance cut in half at each step)).. Obviously, Zeno meant this as a joke. There IS in fact an ABSOLUTE Continuum, what Aristotle called "Being-In-Itself", the Ousia of the Stoics, the "One" of Plotinus, the Tao, the Ein Sof of the Zohar, the Sat-Chit-Ananda of Shankara, and the Rigpa of Buddhism. This is experiential but not in a dualistic sense. Access "Mahamritunjaya Mantra - Sacred Sounds Choir" and listen to it for 5 min per day for at least two weeks. In due time your mind will transcend itself (demolishing the idea of a separate observer), and AHA, what's left is Pure Consciousness, In-Itself, the Absolute Continuum.