Even the figures in qualitative research papers have this energy.

The YouTube shorts algorithm is almost godly. I gave my nephew the YouTube app in incognito mode and within 2 mins, it was feeding him exactly what he usually watches on our home TV. AI researchers should focus on explainable agents and quit the AGI race. AI is too good already.

Why hide your impostor syndrome when you could publish it in the Journal of Econometrics and win a Nobel Prize😁(Clive Granger, 1988)

New preprint! “Collective behavior from surprise minimization,” by me, Beren Millidge, Lancelot Da Costa, Richard Mann, Karl Friston, Iain Couzin
Paper: arxiv.org/abs/2307.14804
Code: github.com/conorheins/col…
🧵 (1/14)

Amazing work, Tamrin Barta! Way to go 👏...

A short blog post from VERSES AI Research about upcoming 2024 benchmark work from VERSES: a discussion of what to expect and when—and why we chose the benchmarks that we did
verses.ai/rd-blog/on-upc…

Maxwell Ramstead VERSES AI Research Tommaso Salvatori Alec Tschantz q(Alex Kiefer | everything else) MKoudahl Aswin Paul Chris L Buckley Takuya Isomura Adeel Razi Brett J. Kagan Congratulations, hard to keep up with you guys! 😂

Maxwell Ramstead VERSES AI Research Tommaso Salvatori Alec Tschantz q(Alex Kiefer | everything else) MKoudahl Aswin Paul Chris L Buckley Takuya Isomura Adeel Razi Brett J. Kagan Great work! And a great thread summarizing it! I'm very interested to look for the contrast against RL

It's all linear algebra

Maxwell Ramstead VERSES AI Research Tommaso Salvatori Alec Tschantz q(Alex Kiefer | everything else) MKoudahl Aswin Paul Chris L Buckley Takuya Isomura Adeel Razi Brett J. Kagan Nice work!

New from VERSES AI Research and collaborators: introducing 'inductive planning,' a new form of active inference allowing us to generate intelligent agents that aim to achieve explicit goals

Delighted to see our paper (On efficient computation in active inference arxiv.org/abs/2307.00504) acknowledged in the best way through this amazing work! I'm sure inductive inference is going to take off real fast in the community. Adeel Razi Lancelot Da Costa Noor Sajid

Maxwell Ramstead VERSES AI Research Tommaso Salvatori Alec Tschantz q(Alex Kiefer | everything else) MKoudahl Aswin Paul Chris L Buckley Takuya Isomura Adeel Razi Brett J. Kagan Fantastic paper. It's a great overview of all the key components of Active Inference. 👏

Preprint alert on our new (active inference) collabroation! Follow the thread 👇👇

You may also want to read this paper in combination with our earlier preprint on efficent copmputation of large planning/polcies spaces from Aswin Paul here: arxiv.org/abs/2307.00504

New formulation of active inference just dropped: meet inductive planning

🚨 Preprint 🚨 “Active inference and intentional behaviour,” from VERSES AI Research (Friston, Tommaso Salvatori, Alec Tschantz, q(Alex Kiefer | everything else), Verbelen, MKoudahl, Aswin Paul, Chris L Buckley, me) and collaborators (@TakuyaIsomura, Parr, Adeel Razi, Brett J. Kagan)
arxiv.org/abs/2312.07547 1/8

New paper in NeurIPS in which we invesitgate the dynamics of neural connectivity as the biological cells (aka DishBrain) learn to play Pong.

Link: openreview.net/forum?id=dq53F…

Unfortunately Moein Khajehnejad & Forough Habibollahi who led the work cannot get US visa to go & present in person

Once more, NeurIPS Conference is in US, and once again, we can't attend due to visa processing delays! If you're there, please join our virtual oral presentation on Dec 16th at the 'Symmetry and Geometry in Neural Representations' workshop;

Time is a scam. 😭