Sign in

Takuya Isomura

@takuyaisomura.bsky.social
29 followers 6 following 6 posts

Theoretical neuroscientist. Unit Leader (PI), Brain Intelligence Theory Unit, RIKEN CBS. I'm interested in universal characterisation of the brain intelligence. cbs.riken.jp/en/faculty/t.isomura

PostsRepliesMedia
Takuya Isomura @takuyaisomura.bsky.social · 26/11/2025
Congrats to Ukyo Tazawa for publishing his first first-author paper in Neural Networks!🎉 He develops a new Bayesian model selection method based on a synaptic pruning rule derived from the variational principle. #ActiveInference #FreeEnergyPrinciple doi.org/10.1016/j.ne...
doi.org
Redirecting
020
Takuya Isomura @takuyaisomura.bsky.social · 18/10/2025
Got the Best Poster Award at #IWAI2025! Many thanks to everyone who listened to my spotlight/stopped by my poster and shared thoughtful questions & comments! IWAI 2025▶️ iwaiworkshop.github.io Related preprint▶️ doi.org/10.1101/2025...
010
Takuya Isomura @takuyaisomura.bsky.social · 11/08/2025
Our new preprint on active inference in zebrafish, with Yuki Tanimoto, Makio Torigoe, Hitoshi Okamoto, and Hideaki Shimazaki "Predicting individual learning trajectories in zebrafish via the free-energy principle" doi.org/10.1101/2025...
doi.org
Predicting individual learning trajectories in zebrafish via the free-energy principle
The free-energy principle has been proposed as a unified theory of brain function, and recent evidence from in vitro experiments supports its validity. However, its empirical application to in vivo ne...
421
Reposted by Takuya Isomura
bioRxiv Neuroscience @biorxiv-neursci.bsky.social · 07/08/2025
Predicting individual learning trajectories in zebrafish via the free-energy principle www.biorxiv.org/content/10.1101/202…
031
Takuya Isomura @takuyaisomura.bsky.social · 15/04/2025
Our new paper "Triple equivalence for the emergence of biological intelligence" was published in Communications Physics. www.nature.com/articles/s42... Three pillars of intelligence–dynamical systems of neural nets, statistical inference, and Turing machines–have been studied almost separately. 1/2
nature.com
Triple equivalence for the emergence of biological intelligence - Communications Physics
Characterizing the intelligence of biological organisms is challenging yet crucial. This paper demonstrates the capacity of canonical neural networks to autonomously generate diverse intelligent algor...
110
Takuya Isomura @takuyaisomura.bsky.social · 15/04/2025
hello, world
000