AI Paper Atlas

Research map / Continual learning / Online and replay-based

SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning

Lirui Luo, Guoxi Zhang, Hongming Xu, Cong Fang, Qing Li

ICML 2026 · Continual learning › Online and replay-based · approach: Plasticity-preserving continual reinforcement learning · 0 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
add the SPHERE Parseval penalty on last-layer weighted expert features (ablation)Top-K MoE PPO · HumanoidBench CRLaverage success: 0.36 -> 0.54Improved
add SPHERE to Top-K MoETop-K MoE PPO · MetaWorld CRLaverage success: +133%Improved
add SPHERE to Top-K MoETop-K MoE PPO · HumanoidBench RLaverage success: +36%Improved
add SPHERE to Top-K MoETop-K MoE PPO · HumanoidBench CRLaverage success: +50%Improved
regularize only the last hidden expert layer instead of all hidden expert layers (ablation)SPHERE · HumanoidBench CRLaverage success: 0.42 -> 0.54Improved
add cross-expert concatenation instead of per-expert loss sum (ablation)SPHERE · HumanoidBench CRLaverage success: 0.40 -> 0.54Improved
regularize feature Gram A_last instead of gradient Gram G_last (ablation)SPHERE · HumanoidBench CRLaverage success: 0.43 -> 0.54Improved
switch from RL to CRLPPO and MoE policies · HumanoidBenchaverage successHurt

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