SafeAdapt certifies a Rashomon set of safe policies from demonstration data and projects updates from arbitrary RL algorithms onto it to guarantee preservation of safety on source tasks.
11 Preprint
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
While deep learning has led to remarkable advances across diverse applications, it struggles in domains where the data distribution changes over the course of learning. In stark contrast, biological neural networks continually adapt to changing domains, possibly by leveraging complex molecular machinery to solve many tasks simultaneously. In this study, we introduce intelligent synapses that bring some of this biological complexity into artificial neural networks. Each synapse accumulates task relevant information over time, and exploits this information to rapidly store new memories without forgetting old ones. We evaluate our approach on continual learning of classification tasks, and show that it dramatically reduces forgetting while maintaining computational efficiency.
fields
cs.LG 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
DG-Hard uses Donoho-Gavish hard thresholding on the fine-tuning weight delta to separate task-aligned signal from noise-like residual, recovering damaged capabilities while preserving target-task gains.
The relative rankings of continual learning methods are not preserved across different fine-tuning regimes defined by trainable parameter depth.
citing papers explorer
-
SafeAdapt: Provably Safe Policy Updates in Deep Reinforcement Learning
SafeAdapt certifies a Rashomon set of safe policies from demonstration data and projects updates from arbitrary RL algorithms onto it to guarantee preservation of safety on source tasks.
-
Spectral Unforgetting: Post-Hoc Recovery of Damaged Capabilities Without Retraining
DG-Hard uses Donoho-Gavish hard thresholding on the fine-tuning weight delta to separate task-aligned signal from noise-like residual, recovering damaged capabilities while preserving target-task gains.
-
Fine-Tuning Regimes Define Distinct Continual Learning Problems
The relative rankings of continual learning methods are not preserved across different fine-tuning regimes defined by trainable parameter depth.