Replay Can Provably Increase Forgetting

• URL: http://arxiv.org/abs/2506.04377v1
• Date: Wed, 04 Jun 2025 18:46:23 GMT
• Title: Replay Can Provably Increase Forgetting
• Authors: Yasaman Mahdaviyeh, James Lucas, Mengye Ren, Andreas S. Tolias, Richard Zemel, Toniann Pitassi,
• Abstract summary: A critical challenge for continual learning is forgetting, where the performance on previously learned tasks decreases as new tasks are introduced.<n>One of the commonly used techniques to mitigate forgetting, sample replay, has been shown empirically to reduce forgetting.<n>We show that even in a noiseless setting, forgetting can be non-monotonic with respect to the number of replay samples.
• Score: 24.538643224479515
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Continual learning seeks to enable machine learning systems to solve an increasing corpus of tasks sequentially. A critical challenge for continual learning is forgetting, where the performance on previously learned tasks decreases as new tasks are introduced. One of the commonly used techniques to mitigate forgetting, sample replay, has been shown empirically to reduce forgetting by retaining some examples from old tasks and including them in new training episodes. In this work, we provide a theoretical analysis of sample replay in an over-parameterized continual linear regression setting, where each task is given by a linear subspace and with enough replay samples, one would be able to eliminate forgetting. Our analysis focuses on sample replay and highlights the role of the replayed samples and the relationship between task subspaces. Surprisingly, we find that, even in a noiseless setting, forgetting can be non-monotonic with respect to the number of replay samples. We present tasks where replay can be harmful with respect to worst-case settings, and also in distributional settings where replay of randomly selected samples increases forgetting in expectation. We also give empirical evidence that harmful replay is not limited to training with linear models by showing similar behavior for a neural networks equipped with SGD. Through experiments on a commonly used benchmark, we provide additional evidence that, even in seemingly benign scenarios, performance of the replay heavily depends on the choice of replay samples and the relationship between tasks.

Related papers

• Non-Uniform Memory Sampling in Experience Replay [1.9580473532948401]
  A popular strategy to alleviate catastrophic forgetting is experience replay.<n>Most approaches assume that sampling from this buffer is uniform by default.<n>We generate 50 different non-uniform sampling probability weights for each trial and compare their final accuracy to the uniform sampling baseline.
  arXiv  Detail & Related papers  (2025-02-16T23:04:16Z)
• Continual Diffuser (CoD): Mastering Continual Offline Reinforcement Learning with Experience Rehearsal [54.93261535899478]
  In real-world applications, such as robotic control of reinforcement learning, the tasks are changing, and new tasks arise in a sequential order.<n>This situation poses the new challenge of plasticity-stability trade-off for training an agent who can adapt to task changes and retain acquired knowledge.<n>We propose a rehearsal-based continual diffusion model, called Continual diffuser (CoD), to endow the diffuser with the capabilities of quick adaptation (plasticity) and lasting retention (stability)
  arXiv  Detail & Related papers  (2024-09-04T08:21:47Z)
• Watch Your Step: Optimal Retrieval for Continual Learning at Scale [1.7265013728931]
  In continual learning, a model learns incrementally over time while minimizing interference between old and new tasks. One of the most widely used approaches in continual learning is referred to as replay. We propose a framework for evaluating selective retrieval strategies, categorized by simple, independent class- and sample-selective primitives. We propose a set of strategies to prevent duplicate replays and explore whether new samples with low loss values can be learned without replay.
  arXiv  Detail & Related papers  (2024-04-16T17:35:35Z)
• Enhancing Consistency and Mitigating Bias: A Data Replay Approach for Incremental Learning [93.90047628101155]
  Deep learning systems are prone to catastrophic forgetting when learning from a sequence of tasks.<n>To address this, some methods propose replaying data from previous tasks during new task learning.<n>However, it is not expected in practice due to memory constraints and data privacy issues.
  arXiv  Detail & Related papers  (2024-01-12T12:51:12Z)
• Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study [68.75670223005716]
  We find that pre-trained language models like BERT have a potential ability to learn sequentially, even without any sparse memory replay. Our experiments reveal that BERT can actually generate high quality representations for previously learned tasks in a long term, under extremely sparse replay or even no replay.
  arXiv  Detail & Related papers  (2023-03-02T09:03:43Z)
• Reducing Training Sample Memorization in GANs by Training with Memorization Rejection [80.0916819303573]
  We propose rejection memorization, a training scheme that rejects generated samples that are near-duplicates of training samples during training. Our scheme is simple, generic and can be directly applied to any GAN architecture.
  arXiv  Detail & Related papers  (2022-10-21T20:17:50Z)
• An Investigation of Replay-based Approaches for Continual Learning [79.0660895390689]
  Continual learning (CL) is a major challenge of machine learning (ML) and describes the ability to learn several tasks sequentially without catastrophic forgetting (CF) Several solution classes have been proposed, of which so-called replay-based approaches seem very promising due to their simplicity and robustness. We empirically investigate replay-based approaches of continual learning and assess their potential for applications.
  arXiv  Detail & Related papers  (2021-08-15T15:05:02Z)
• Distilled Replay: Overcoming Forgetting through Synthetic Samples [11.240947363668242]
  Replay strategies are Continual Learning techniques which mitigate catastrophic forgetting by keeping a buffer of patterns from previous experience. This work introduces Distilled Replay, a novel replay strategy for Continual Learning which is able to mitigate forgetting by keeping a very small buffer. We show the effectiveness of our Distilled Replay against naive replay, which randomly samples patterns from the dataset, on four popular Continual Learning benchmarks.
  arXiv  Detail & Related papers  (2021-03-29T18:02:05Z)
• Selective Replay Enhances Learning in Online Continual Analogical Reasoning [44.794321821598395]
  In continual learning, a system learns from non-stationary data streams or batches without catastrophic forgetting. While this problem has been heavily studied in supervised image classification and reinforcement learning, continual learning in neural networks designed for abstract reasoning has not yet been studied.
  arXiv  Detail & Related papers  (2021-03-06T00:04:10Z)
• Stratified Experience Replay: Correcting Multiplicity Bias in Off-Policy Reinforcement Learning [17.3794999533024]
  We show that deep RL appears to struggle in the presence of extraneous data. Recent works have shown that the performance of Deep Q-Network (DQN) degrades when its replay memory becomes too large. We re-examine the motivation for sampling uniformly over a replay memory, and find that it may be flawed when using function approximation.
  arXiv  Detail & Related papers  (2021-02-22T19:29:18Z)
• Automatic Recall Machines: Internal Replay, Continual Learning and the Brain [104.38824285741248]
  Replay in neural networks involves training on sequential data with memorized samples, which counteracts forgetting of previous behavior caused by non-stationarity. We present a method where these auxiliary samples are generated on the fly, given only the model that is being trained for the assessed objective. Instead the implicit memory of learned samples within the assessed model itself is exploited.
  arXiv  Detail & Related papers  (2020-06-22T15:07:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.