@phanseneecs Profile picture

Philippe Hansen

@phanseneecs

Researching Deep Reinforcement Learning at BAIR, UC Berkeley

Similar User
Kevin Black photo

@kvablack

Cassidy Laidlaw photo

@cassidy_laidlaw

Youngwoon Lee photo

@YoungwoonLee

Xinyang (Young) Geng photo

@younggeng

Seohong Park photo

@seohong_park

Brandon Trabucco photo

@brandontrabucco

Soroush Nasiriany photo

@snasiriany

Kimin photo

@kimin_le2

Partha Ghosh photo

@partha_ghosh0

Boyuan Chen photo

@BoyuanChen0

Younggyo Seo photo

@younggyoseo

Katie Kang photo

@katie_kang_

Philipp Wu photo

@philippswu

Mitsuhiko Nakamoto photo

@mitsuhiko_nm

Qiyang Li photo

@qiyang_li

Philippe Hansen Reposted

Building generalizable goal-conditioned agents from rich observations is a key to solving real world problems. We propose a new form of state abstraction, goal-conditioned bisimulation, that captures functional equivariance. 1/n Paper: arxiv.org/abs/2204.13060


Loading...

Something went wrong.


Something went wrong.