AI Paper Atlas

Research map / Continual learning / Online and replay-based

Resolving the Stability-Plasticity Dilemma in Reinforcement Learning via Complementary Continual Critics

Bo Sun, Peixi Peng, Guang Tan, Haoran Xu, Yaokun Li, Yiqian Chang, Shuaixian Wang, Luntong Li

CVPR 2026 · Continual learning › Online and replay-based · approach: Plasticity-preserving continual reinforcement learning

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
apply CBP to one critic only or EWC to one critic only (ablation)CD-CCA · Quadruped Walk and CARLA Highwaytraining returnImproved
replace cross-attention fusion with minimum of two critics (ablation)CD-CCA · Quadruped Walk and CARLA Highwaytraining returnHurt
use cross-attention fusion without CBP and EWC (ablation)SAC baseline · Quadruped Walk and CARLA Highwaysample efficiencyImproved
use two CBP critics instead of CBP plus EWC (ablation)CD-CCA · Quadruped Walk and CARLA Highwaytraining returnHurt
use two EWC critics instead of CBP plus EWC (ablation)CD-CCA · Quadruped Walk and CARLA Highwaytraining returnHurt
extend SAC and TD3 to four critics (two CBP, two EWC) (ablation)SAC, TD3 · Quadruped Walk and CARLA HighwayconvergenceImproved
add CD-CCA to DrQ-v2DrQ-v2 · DMControl 4 hard tasks (500k)average episode return: 636.8 -> 683.0Improved
add CD-CCA to DrQ-v2DrQ-v2 · DMControl Pendulum Swingup (500k)episode return: 842±12 -> 756±119Hurt
add CD-CCA to DrQ-v2 at 1M stepsDrQ-v2 · DMControl Pendulum Swingup (1M)episode return: 812±23 -> 817±17No clear effect
add CD-CCA to DeepMDPDeepMDP · CARLA Highwayepisode reward: 170±36 -> 343±63Improved
add CD-CCA to DeepMDPDeepMDP · CARLA Highwaydriving distance (m): 132±20 -> 287±45 (ResAct 299±24 still higher)Mixed

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