AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

DOMINO: Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning

Yao Mu, Yuzheng Zhuang, Fei Ni, Bin Wang, Jianyu Chen, Jianye HAO, Ping Luo

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: Task-inference meta-RL · 2 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Cartpole-f-ltest average return: 234 -> 283Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Pendulum-m-ltest average return: -526 -> -436Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Ant-m-dtest average return: 184 -> 216Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Halfcheetah-m-dtest average return: 824 -> 1034Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Slimhumanoid-m-dtest average return: 4275 -> 5258Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Hopper-m-dtest average return: 964 -> 1137Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Cripple-Ant-m-dtest average return: 113 -> 132Improved
replace entangled context (MINO) with decomposed MI optimization (DOMINO) (ablation)PPO+DOMINO · Cripple-Halfcheetah-m-dtest average return: 1197 -> 1326Improved
use DOMINO context instead of PEARL contextPPO+DOMINO · Hopper-m-dtest average return: 874 -> 1137Improved
use DOMINO context instead of T-MCL contextPPO+DOMINO · Ant-m-dtest average return: 173 -> 216Improved
use DOMINO context instead of T-MCL contextPPO+DOMINO · Slimhumanoid-m-dtest average return: 4136 -> 5258Improved
PPO+DOMINO vs PEARL on training returnPPO+DOMINO · Cripple-Halfcheetah-m-dtest average return: 2538 -> 2503No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy