I've nothing to contribute intellectually but I must confess that the current furore over consciousness theories brings to mind an image of a field full of people on various-sized stepladders all arguing loudly about who's closest to the moon

One of the biggest problems with LLM is that, after a while, they loose the thread Self-organizing systems (like human society and muluticellular organisnms) don't have this problem. They can scale in size as much as they want. Is it possible to make a self-organizing LLM? 🧵

New preprint: francesco215.github.io/Language_CA/, Francesco Sacco & dalton, in 3 versions: for biologists, AI folks, & physicists. x.com/FrancescoSacco… Self-organizing systems' topology & linguistic space navigation by LLM: what limits scaling, keeping a longer narrative going?

Conversation with Dr. Jeff Beck from @versesai on #activeinference the Bayesian Brain, emergence and optimisation. youtube.com/watch?v=c4praC…

Brilliant conversation with Lancelot Da Costa about active inference, multi agent learning, core knowledge, structure learning and intelligence. youtube.com/watch?v=qxEfcr…

Mathematician Doug Ravenel’s seven conjectures in topology had all proven to be true, save one. That conjecture was found to be false earlier this year, and upends much of what mathematicians thought about the relationship between shapes and dimensions. quantamagazine.org/an-old-conject…

Really excited about this paper, featuring brand new PhD Gizem Gumuskaya 🎉, collaborator @sjmgarnier, & undergrad students P. Srivastava, B. Cooper, H. Lesser, B. Semegran: onlinelibrary.wiley.com/doi/10.1002/ad… What I personally think it means: thoughtforms.life/meet-the-anthr…

Just published a video explanation on the role self-organization plays in modern language models! hope you like it: youtu.be/cGcY-ReeGDU

Apocalyptic Brexit was always the wrong call - the boring truth is we're just gonna be a bit poorer forever. Mundane misery. It obviously wouldn't resolve the social contradictions its fantasist supporters hoped for, and we obviously were never gonna be Singapore-on-the-Atlantic.

My brilliant trainee Piotr Sokol (Piotr Sokol) PhD is on the job market! He's an expert in machine learning, differential geometry, dynamical systems / stochastic processes, and lover of manifolds and algebra.

📜✨ New paper out! In this jww Owen Lynch and David Spivak we introduce 'open energy-driven systems', which are a broad class of differential systems which evolve by generalised gradient flow

RIP Jim Simons

My book with Scott Hotton on Braitenberg Vehicles was published today 🥳. The result of thousands of backyard meetings and zoom calls, the book confirms what I'll call Braitenberg's dictum: even the simplest systems contain great wealths of complexity. 1/

Everyone is mocking this but I can't even bring myself to laugh cause there's something deeply depressing about how my field has been coopted by these dweebs trying to signal intelligence

Check out our new preprint! arxiv.org/abs/2410.02543 We find that Diffusion Models are Evolutionary Algorithms! By viewing evolution as denoising, we show they share the same mathematical foundation. We then propose Diffusion Evolution (1/n)

this is the guy everyone saw burn and melt alive on a hospital bed with an IV attached to him. remember these people are more than just numbers. they had a life, a family, dreams just like us.

Re: that whole Oxford situation, I also was an international student who did a PhD in the UK. If you make it to all the way to 4th year before finding out that your thesis isn’t viable, your supervisors have failed you, likely because the uni wanted to keep extracting £££

New #preprint from Francesco Sacco and dalton - physics, #self-organization, #diverseintelligence: arxiv.org/abs/2501.13188 "Topological constraints on self-organisation in locally interacting systems" Abstract: All intelligence is collective intelligence, in the

After 12 years, my book “Gaming Cancer” is out. Its basic premise is that by embedding scientific problems into the mechanics of different kinds of games (from real-time strategies to first-person shooters) we increase our chances of finding cures to diseases like cancer.