Hybrid Memory Replay: Blending Real and Distilled Data for Class Incremental Learning

• URL: http://arxiv.org/abs/2410.15372v1
• Date: Sun, 20 Oct 2024 12:13:32 GMT
• Title: Hybrid Memory Replay: Blending Real and Distilled Data for Class Incremental Learning
• Authors: Jiangtao Kong, Jiacheng Shi, Ashley Gao, Shaohan Hu, Tianyi Zhou, Huajie Shao,
• Abstract summary: Incremental learning (IL) aims to acquire new knowledge from current tasks while retaining knowledge learned from previous tasks. Replay-based IL methods store a set of exemplars from previous tasks in a buffer and replay them when learning new tasks. Data distillation (DD) can reduce the exemplar buffer's size, by condensing a large real dataset into a much smaller set of more information-compact synthetic exemplars. We propose an innovative modification to DD that distills synthetic data from a sliding window of checkpoints in history.
• Score: 19.671792951465
• License:
• Abstract: Incremental learning (IL) aims to acquire new knowledge from current tasks while retaining knowledge learned from previous tasks. Replay-based IL methods store a set of exemplars from previous tasks in a buffer and replay them when learning new tasks. However, there is usually a size-limited buffer that cannot store adequate real exemplars to retain the knowledge of previous tasks. In contrast, data distillation (DD) can reduce the exemplar buffer's size, by condensing a large real dataset into a much smaller set of more information-compact synthetic exemplars. Nevertheless, DD's performance gain on IL quickly vanishes as the number of synthetic exemplars grows. To overcome the weaknesses of real-data and synthetic-data buffers, we instead optimize a hybrid memory including both types of data. Specifically, we propose an innovative modification to DD that distills synthetic data from a sliding window of checkpoints in history (rather than checkpoints on multiple training trajectories). Conditioned on the synthetic data, we then optimize the selection of real exemplars to provide complementary improvement to the DD objective. The optimized hybrid memory combines the strengths of synthetic and real exemplars, effectively mitigating catastrophic forgetting in Class IL (CIL) when the buffer size for exemplars is limited. Notably, our method can be seamlessly integrated into most existing replay-based CIL models. Extensive experiments across multiple benchmarks demonstrate that our method significantly outperforms existing replay-based baselines.

Related papers

• F-OAL: Forward-only Online Analytic Learning with Fast Training and Low Memory Footprint in Class Incremental Learning [28.772554281694166]
  Online Class Incremental Learning (OCIL) aims to train models incrementally, where data arrive in mini-batches, and previous data are not accessible. A major challenge in OCIL is Catastrophic Forgetting, i.e., the loss of previously learned knowledge. We propose an exemplar-free approach--Forward-only Online Analytic Learning (F-OAL) Unlike traditional methods, F-OAL does not rely on back-propagation and is forward-only, significantly reducing memory usage and computational time.
  arXiv  Detail & Related papers  (2024-03-23T07:39:13Z)
• Enhancing Consistency and Mitigating Bias: A Data Replay Approach for Incremental Learning [100.7407460674153]
  Deep learning systems are prone to catastrophic forgetting when learning from a sequence of tasks. To mitigate the problem, a line of methods propose to replay the data of experienced tasks when learning new tasks. However, it is not expected in practice considering the memory constraint or data privacy issue. As a replacement, data-free data replay methods are proposed by inverting samples from the classification model.
  arXiv  Detail & Related papers  (2024-01-12T12:51:12Z)
• A Memory Transformer Network for Incremental Learning [64.0410375349852]
  We study class-incremental learning, a training setup in which new classes of data are observed over time for the model to learn from. Despite the straightforward problem formulation, the naive application of classification models to class-incremental learning results in the "catastrophic forgetting" of previously seen classes. One of the most successful existing methods has been the use of a memory of exemplars, which overcomes the issue of catastrophic forgetting by saving a subset of past data into a memory bank and utilizing it to prevent forgetting when training future tasks.
  arXiv  Detail & Related papers  (2022-10-10T08:27:28Z)
• Few-Shot Class-Incremental Learning via Entropy-Regularized Data-Free Replay [52.251188477192336]
  Few-shot class-incremental learning (FSCIL) has been proposed aiming to enable a deep learning system to incrementally learn new classes with limited data. We show through empirical results that adopting the data replay is surprisingly favorable. We propose using data-free replay that can synthesize data by a generator without accessing real data.
  arXiv  Detail & Related papers  (2022-07-22T17:30:51Z)
• Memory Replay with Data Compression for Continual Learning [80.95444077825852]
  We propose memory replay with data compression to reduce the storage cost of old training samples. We extensively validate this across several benchmarks of class-incremental learning and in a realistic scenario of object detection for autonomous driving.
  arXiv  Detail & Related papers  (2022-02-14T10:26:23Z)
• Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning [73.24988226158497]
  We consider the high-impact problem of Data-Free Class-Incremental Learning (DFCIL) We propose a novel incremental distillation strategy for DFCIL, contributing a modified cross-entropy training and importance-weighted feature distillation. Our method results in up to a 25.1% increase in final task accuracy (absolute difference) compared to SOTA DFCIL methods for common class-incremental benchmarks.
  arXiv  Detail & Related papers  (2021-06-17T17:56:08Z)
• ZS-IL: Looking Back on Learned ExperiencesFor Zero-Shot Incremental Learning [9.530976792843495]
  We propose an on-call transfer set to provide past experiences whenever a new class arises in the data stream. ZS-IL demonstrates significantly better performance on the well-known datasets (CIFAR-10, Tiny-ImageNet) in both Task-IL and Class-IL settings.
  arXiv  Detail & Related papers  (2021-03-22T22:43:20Z)
• Improving Computational Efficiency in Visual Reinforcement Learning via Stored Embeddings [89.63764845984076]
  We present Stored Embeddings for Efficient Reinforcement Learning (SEER) SEER is a simple modification of existing off-policy deep reinforcement learning methods. We show that SEER does not degrade the performance of RLizable agents while significantly saving computation and memory.
  arXiv  Detail & Related papers  (2021-03-04T08:14:10Z)
• Condensed Composite Memory Continual Learning [17.192367229752072]
  Deep Neural Networks (DNNs) suffer from a rapid decrease in performance when trained on a sequence of tasks where only data of the most recent task is available. We propose a novel way of learning a small set of synthetic examples which capture the essence of a complete dataset.
  arXiv  Detail & Related papers  (2021-02-19T12:18:15Z)

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.