I have no words!!! I am beyond grateful to have been announced as one of this year’s Glushko prize winners.

Thank you to CogSci Society and Bob Glushko for supporting this award. HUGE thank you to my PhD advisors: Chris Baker and Plasticity Lab

Our paper 'Collective behavior from surprise minimization' now published in PNAS PNASNews! Check it out to learn about a new, flexible approach to modelling collective motion with active inference!

Minoritised communities are underrepresented in UK research – particularly those of Black, Bangladeshi or Pakistani heritage.

We've launched new funding to help address this ⤵️ [1/3]

If you are in nyc this Friday come along to the opening of Jimmy Beauquesne's show Undawn where I will be giving a talk on my PhD work inspired by the pieces in the show ✨

fragment.gallery/exhibitions/65…

What governs the temporal variability of cognition? Grateful to see our work now published in Nature Human Behaviour with Daniel Reznik Rahul Bhui, Hanna Hillman and Sam Gershman. Read all about it: nature.com/articles/s4156…

Please take note: VERSES AI Research is hiring
apply.workable.com/verses/

How is active inference a model of agency? How does it relate to RL?
+ AIF and expected free energy derived
+ the generic AIF algorithm (no one implements this yet! 😎)
+ learning preferences 😮
Arxiv from our 2022 RL and Decision Making Conference paper! arxiv.org/pdf/2401.12917…

Here is a YouTube link to my #NeurIPS2023 invited talk 'Open-ended and AI-generating Algorithms in the Era of Foundation Models.' Includes exciting new OMNI results in an infinite task space. Lmkwyt! youtu.be/TO_xQzWXTcI NeurIPS Conference ALOE Workshop Goal-Conditioned RL Workshop Joseph Suarez (e/🐡)

So excited to be part of the Intrinsically Motivated Open-Ended Learning workshop at #NeurIPS2023 ! First up: Doina Precup sketches out a general blueprint for continual reinforcement learning.

More details on today’s workshop here: imol-workshop.github.io

If you're at NeurIPS, swing by Poster #1226 at 17:15 to discuss our paper on applying SSM's (like Mamba) to RL!

A bonus: We've released a pure JAX repo for testing these architectures that includes 34 memory-based environments and runs 100x faster

github.com/luchris429/pop…

How can animals learn from rewards across timescales? Our work with Pablo Tano HyungGoo Kim Athar N Malik, MD, PhD Alexandre Pouget Naoshige Uchida explores how dopamine neurons could enable multi-timescale reinforcement learning in the brain
biorxiv.org/content/10.110…

I’ve seen some talk recently about the virtues of hallucination, implying that it’s a “good” thing because it’s “creative” to make things up or simply what LMs are “trained to do.” I see my own work sometimes come up in these discussions since I’ve written a lot on creativity

🚨 Job alert!
The Gatsby Unit invites applications for an open-rank faculty position in #MachineLearning .
ℹ️ ucl.ac.uk/gatsby/vacanci…
⏰ Review begins 9 Jan 2024; apply by 8 Jan for full consideration
👋 Chat with Arthur Gretton, Maneesh Sahani or Andrew Saxe at #NeurIPS2023

📣Our 🆕 paper Causal Inference is Not Just a Statistics Problem is out! Malco`lm()` Barrett (Now on Bluesky and Mastodon), King Ranch @[email protected], and I show that you can have 4 data sets with identical summary stats & visuals but very different data generating mechanisms-statistics alone can't tell you what to adjust for!

Does the success of deep neural networks in creating AI systems mean Bayesian models are no longer relevant? Our new paper argues the opposite: these approaches are complementary, creating new opportunities to use Bayes to understand intelligent machines
arxiv.org/abs/2311.10206

New episode! ✨ This week we spoke to Anil Seth about all things #consciousness ! From controlled hallucinations to implications for #AI Anil shares with us his ideas on where the field is currently at 🤔🧠

open.spotify.com/episode/2FJCmR…

Generating meaning: active inference and the scope and limits of passive AI

Opinion by Giovanni Pezzulo, Thomas Parr, Paul Cisek, Andy Clark, & Karl Friston
OA: doi.org/10.1016/j.tics…

What a lovely way to wrap up my time in Melbourne - with fantastic tutorial sessions from Johan Medrano and Aswin Paul! Thank you Jakob Hohwy, Beth Fisher and co for hosting this! 🎉

What do we talk about when we talk about 'curiosity'? 🤔
In our new paper in Trends in Neurosciences (with Kacper Kondrakiewicz, Gerstner Lab & Sebastian Haesler), we address this question by reviewing the behavioral signatures, neural mechanisms, and comp. models of curiosity:
doi.org/10.1016/j.tins…

Our new preprint led by the great Karl Friston, in which we explore how the structure of generative models can be learnt from data

arxiv.org/abs/2311.10300