AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Self-Supervised Mixture-of-Experts by Uncertainty Estimation

Zhuobin Zheng, Chun Yuan, Xinrui Zhu, Zhihui Lin, Yangyang Cheng, Cheng Shi, Jiahui Ye

AAAI 2019 · Reinforcement learning › Visual and representation RL · approach: In-context and adaptive RL · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah SmallFoot (fully trained)cumulative reward: 2003.46 -> 3460Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah BigFoot (fully trained)cumulative reward: 2211.92 -> 2392Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah SmallLeg (fully trained)cumulative reward: 2327.16 -> 4233Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah BigLeg (fully trained)cumulative reward: 2269.78 -> 3147Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah SmallThigh (fully trained)cumulative reward: 2555.16 -> 3346Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah BigThigh (fully trained)cumulative reward: 2424.95 -> 2465Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah SmallTorso (fully trained)cumulative reward: 2294.72 -> 2870Improved
use SUM (K=3) instead of TRPOmulti-task agent · HalfCheetah BigTorso (fully trained)cumulative reward: 2686.13 -> 3478Improved
train with NLL uncertainty objective instead of MSE, single head K=1 (ablation)DDPG · Walker and Hopperaveraged returnImproved
add uncertainty enhancement with K=3 heads versus multi-head DDPG (ablation)SUM (K=3) · Walker/Hopperaveraged return and learning speedImproved
uncertainty-trained critic versus multi-head DDPG, generalization gap (ablation)SUM · Hopper and Walker test vs trainperformance decrease factor alpha: [0.32, 0.51] -> [0.05, 0.28]Improved
self-supervised expert scheduling versus SOUP confidence strategySUM · Hopper and Humanoid multi-taskaveraged returnImproved

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