paper-with-me

홈 › Papers

The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RL

2026-07-22 · Gurp Nijjer arxiv

Model-based reinforcement-learning agents of the DreamerV3 family forget catastrophically when trained on task sequences, even when an unbounded replay buffer preserves every earlier experience. We ask a question the continual-RL literature has assumed an answer to but never measured: which component forgets? Under never-clear replay, pre-registered component-level probes (n=3 seeds throughout) show that the world model retains essentially everything measurable about old tasks -- reward discrimination (retention ratio ~1.0), value estimates, and termination structure -- while the actor's behavior collapses. Forgetting in this regime is a channel problem, not a memory problem. We demonstrate this by intervention: with the world model frozen and identical imagined rollouts, reinforcement learning in imagination fails to recover a lost skill (0/3 seeds), while supervised self-imitation on the world model's own graded dreams recovers it on 3/3 seeds with zero environment interaction. Interleaved during training, this graded dream rehearsal yields a task-label-free, parameter-constant continual learner: 3/3 four-task chains retained where plain replay passes 0/3, 3/3 eight-task chains, and consistent gains over matched real-episode cloning (paired difference +0.13, bootstrap 95% CI [0.07, 0.24], complete seed separation). The dream-grading step is load-bearing: we characterize two scoring failure modes, provide an offline selection gauge that caught both before they contaminated results, and give a realized-first grading rule that closes them. All experiments were pre-registered with committed protocols; every refuted hypothesis is reported.

📄 PDF Abstract BibTeX arXiv:2607.19749

Code (4)

Tavish9/awesome-daily-AI-arxiv ★ 112
arxivsub/arXivSub_daily_arxiv ★ 3
gurpnijjer/dream-rehearsal ★ 2
🤗 gurpnijjer/dream-rehearsal

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Dynamic Prototype Rehearsal for Continual Learning in ECG Arrhythmia Detection

2025-01-13 · Sana Rahmani, Reetam Chatterjee, Ali Etemad, Javad Hashemi

Continual Learning (CL) methods aim to learn from a sequence of tasks while avoiding the challenge of forgetting previous knowledge. We present DREAM-CL, a novel CL method for ECG arrhythmia detection that introduces dyn…

Arrhythmia DetectionContinual Learning

Non-parametric Memory for Spatio-Temporal Segmentation of Construction Zones for Self-Driving

2021-01-18 · Min Bai, Shenlong Wang, Kelvin Wong, Ersin Yumer 외

In this paper, we introduce a non-parametric memory representation for spatio-temporal segmentation that captures the local space and time around an autonomous vehicle (AV). Our representation has three important propert…

Pseudo-Recursal: Solving the Catastrophic Forgetting Problem in Deep Neural Networks

2018-02-12 · Craig Atkinson, Brendan McCane, Lech Szymanski, Anthony Robins

In general, neural networks are not currently capable of learning tasks in a sequential fashion. When a novel, unrelated task is learnt by a neural network, it substantially forgets how to solve previously learnt tasks. …

Generative Adversarial Network

Don't Forget the Critic: Value-Based Data Rehearsal for Multi-Cyclic Continual Reinforcement Learning

2026-05-21 · Benjamin Poole, Andrew Quinn, Li Yang, Minwoo Lee arxiv

Data rehearsal has emerged as a leading approach for mitigating catastrophic forgetting in Continual Reinforcement Learning (CRL). However, existing work remains confined to policy gradient frameworks, regularizing only …

Reinforcement Learning

Pseudorehearsal in actor-critic agents

2017-04-17 · Marochko Vladimir, Leonard Johard, Manuel Mazzara

Catastrophic forgetting has a serious impact in reinforcement learning, as the data distribution is generally sparse and non-stationary over time. The purpose of this study is to investigate whether pseudorehearsal can i…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)