Trust Region Continual Learning as an Implicit Meta-Learner
Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems
2026
Abstract
Continual learning aims to acquire tasks sequentially without catastrophic forgetting, yet standard strategies face a core tradeoff: regularization-based methods (e.g., EWC) can overconstrain updates when task optima are weakly overlapping, while replay-based methods can retain performance but drift due to imperfect replay. We study a hybrid perspective: _trust region continual learning_ that combines generative replay with a Fisher-metric trust region constraint. We show that, under local approximations, the resulting update admits a MAML-style interpretation with a single implicit inner step: replay supplies an old-task gradient signal (query-like), while the Fisher-weighted penalty provides an efficient offline curvature shaping (support-like). This yields an emergent meta-learning property in continual learning: the model becomes an initialization that rapidly _re-converges_ to prior task optima after each task transition, without explicitly optimizing a bilevel objective. Empirically, on task-incremental diffusion image generation and continual diffusion-policy control, trust region continual learning achieves the best final performance and retention, and consistently recovers early-task performance faster than EWC, replay, and continual meta-learning baselines.
BibTeX
@inproceedings{wang-neurips-2026,
title = {Trust Region Continual Learning as an Implicit Meta-Learner},
author = {Wang, Zekun and Gupta, Anant and MacLellan, Christopher J.},
booktitle = {Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems},
year = {2026},
}