TeLAPA maintains archives of behaviorally diverse yet competent policies aligned in a shared latent space to preserve plasticity and enable faster recovery after interference in continual reinforcement learning.
Plasticity Loss in Deep Reinforcement Learning: A Survey
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
abstract
Plasticity refers to a network's ability to adapt to changing data distributions, which is crucial for the successful training of deep reinforcement learning agents. Loss of plasticity causes performance plateaus and contributes to scaling failures, overestimation bias, and insufficient exploration. To deepen the understanding of plasticity loss, we propose a unified definition, examine its drivers and pathologies, and organize over 50 mitigation strategies into the first comprehensive taxonomy of the field. Our analysis shows gaps in current evaluation practices and reveals that general regularization techniques often outperform domain-specific interventions. Future research should prioritize understanding the mechanisms underlying plasticity loss.
citation-role summary
background 3
citation-polarity summary
fields
cs.LG 5roles
background 3polarities
background 3representative citing papers
SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
Safe continual RL methods face a fundamental tension between enforcing safety constraints and preventing catastrophic forgetting in non-stationary environments, with regularization providing only partial mitigation.
The paper surveys CRL literature, proposes a taxonomy of methods into four categories based on knowledge storage and transfer, reviews metrics and benchmarks, and outlines challenges and future research directions.
Smooth-Leaky and Randomized Smooth-Leaky activations mitigate loss of plasticity in continual learning by targeting negative-branch shape and saturation behavior.
citing papers explorer
-
Beyond Single-Model Optimization: Preserving Plasticity in Continual Reinforcement Learning
TeLAPA maintains archives of behaviorally diverse yet competent policies aligned in a shared latent space to preserve plasticity and enable faster recovery after interference in continual reinforcement learning.
-
SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning
SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
-
Safe Continual Reinforcement Learning in Non-stationary Environments
Safe continual RL methods face a fundamental tension between enforcing safety constraints and preventing catastrophic forgetting in non-stationary environments, with regularization providing only partial mitigation.
-
A Survey of Continual Reinforcement Learning
The paper surveys CRL literature, proposes a taxonomy of methods into four categories based on knowledge storage and transfer, reviews metrics and benchmarks, and outlines challenges and future research directions.
-
Activation Function Design Sustains Plasticity in Continual Learning
Smooth-Leaky and Randomized Smooth-Leaky activations mitigate loss of plasticity in continual learning by targeting negative-branch shape and saturation behavior.