Retrospective Adversarial Replay for Continual Learning

By Lilly Kumari, Shengjie Wang, Tianyi Zhou, J. Bilmes. 2022

54 citations recorded by the source corpus.

Continual learning is an emerging research challenge in machine learning that addresses the problem where models quickly ﬁt the most recently trained-on data but suffer from catastrophic forgetting of previous data due to distribution shifts — it does this by maintaining a small historical replay buffer in replay-based methods. To avoid these problems, this paper proposes a method, “Retrospective Adversarial Replay (RAR)”, that synthesizes adversarial samples near the forgetting boundary. RAR perturbs a buffered sample towards its nearest neighbor drawn from the current task in a latent representation space. By replaying such samples, we are able to reﬁne the boundary between previous and current tasks, hence combating forgetting and reducing bias towards the current task. To mitigate the severity of a small replay buffer, we develop a novel MixUp-based strategy to increase replay variation by replaying mixed augmentations. Combined with RAR, this achieves a holistic framework that helps to alleviate catastrophic forgetting. We show that this excels on broadly-used benchmarks and outperforms other continual learning baselines especially when only a small buffer is available. We conduct a thorough ablation study over each key component as well as a hyperparameter sensitivity analysis to demonstrate the effectiveness and robustness of RAR.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Meta & Continual Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Meta & Continual Learning

Citations in the corpus

Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning (2024, methodology): “RAR [21] used the pairwise relations between the exemplars and the new samples and perturb the exemplars to obtain samples close to the decision boundaries.”