🚨 Preprint 🚨 “Active inference and intentional behaviour,” from VERSES AI Research (Friston, Tommaso Salvatori, Alec Tschantz, p(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

🚨 Preprint 🚨 Stoked to share “Supervised structure learning,” from VERSES AI Research authors (Friston, Lancelot Da Costa, Alec Tschantz, p(Alex Kiefer | everything else), Tommaso Salvatori, Koudahl, Conor Heins, Verbelen, Chris L Buckley) and collaborators (Neacsu, Sajid, Markovic, Parr)
arxiv.org/abs/2311.10300 1/5

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

Honored to announce that I'll be writing a book on the FEP and active inference for the The MIT Press @mitpress.bsky.social, tentatively entitled: (Artificial) Intelligence from First Principles: Using the Free Energy Principle and Active Inference for the Study and Design of Ecosystems of Intelligence

Extremely excited to share this survey/manifesto on the past, present, and future of predictive coding (PC), written with a STELLAR team:

arxiv.org/pdf/2308.07870…

Find out what predictive coding *really* is, and the role it could play in the future of machine intelligence.

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)

It was a genuine pleasure and a great honour to discuss the free energy principle, philosophy, and the approach to artificial intelligence that we’re developing at VERSES AI Research helloVERSES with the awesome folks at Machine Learning Street Talk, one of my absolute favourite podcasts!

New preprint :)
Predictive coding networks allow us to perform Bayesian inference on continuous state variables. In this work, we go beyond Bayesian inference, and show how to perform interventional and counterfactual inference.

The final project of my PhD is out at last! A comprehensive version of active inference as pure message passing on graphs😎 You want freeform generative models? With epistemics? That are real fast? Direct policy inference instead of rollouts you say? We got em all right here

V1 of this was awesome. I'm absolutely getting this one

Just published in JOSS: 'RxInfer: A Julia package for reactive real-time Bayesian inference' doi.org/10.21105/joss.…

🚨 Our new flagship paper on the free energy principle & Bayesian mechanics has now been published! 🚨
“On Bayesian mechanics: a physics of and by beliefs” by Dalton A R Sakthivadivel Conor Heins MKoudahl Beren Millidge Lancelot Da Costa Brennan Klein and Karl Friston 1/7
royalsocietypublishing.org/doi/10.1098/rs…

1/15
I should say a bit more about this talk by Benjamin skjold for #EE2023 ee2023.rsvpify.com

It's part of a proper science thriller. Real science with ups and downs and doubts and theory, experiment, empiricism, exhausting arguments, and sleepless nights.

🚨 New preprint 🚨 “Path integrals, particular kinds, and strange things,” on the path integral formulation of the FEP, and kinds of particles—written with Karl Friston, Lancelot Da Costa, Dalton A R Sakthivadivel, Conor Heins, Grigorios Pavliotis, Thomas Parr
arxiv.org/abs/2210.12761 1/18

VERSES Data Science is hiring!

VERSES Research and Development team is headed to Grenoble, France, for the International Workshop on Active Inference > Presenting 9 research papers on active inference. Come for the Alps ⛰️. Stay for the Active Inference🧠.

Learn more >> hubs.li/Q01mrg6N0

I'm quite happy about this new preprint, which begins the development of a semi-axiomatic meta-theory and philosophy for the application and scope of the free-energy principle, and which responds to arguments that FEP-theoretic modelling commits the map-territory fallacy