Home
I explore the nature of intelligence using both biological systems and machine learning approaches. I believe that intelligence can emerge with properties common to the biological nervous system and computers.
In vitro neural networks perform blind source separation by following the free-energy principle
(Isomura et al., 2015, Isomura, Friston, 2018, Spotlight: PLoS Blogs, Gizmodo)
Takuya Isomura Ph.D.
Unit Leader
Brain Intelligence Theory Unit, RIKEN Center for Brain Science
Adjunct Associate Professor
Graduate School of Informatics, Kyoto University
磯村 拓哉（いそむら たくや）
理化学研究所 脳神経科学研究センター
脳型知能理論研究ユニット ユニットリーダー
京都大学 大学院情報学研究科 システム科学専攻 連携准教授
博士（科学）
Error-gated Hebbian plasticity (EGHR): a biologically plausible learning rule for blind source separation
- EGHR for ICA (Isomura, Toyoizumi, 2016, 日本語プレスリリース)
- EGHR for PCA (Isomura, Toyoizumi, 2018, 日本語プレスリリース)
- EGHR for multi-context blind source separation (Isomura, Toyoizumi, 2019)
