Abstract:Deep learning systems are prone to catastrophic forgetting when learning from a sequence of tasks, where old data from experienced tasks is unavailable when learning from a new task. To mitigate the problem, a line of methods propose to replay the data of experienced tasks when learning new tasks. These methods usually adopt an extra memory to store the data for replay. 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. Though achieving good results, these methods still suffer from the inconsistency of the inverted and real training data, which is neglected in the inversion stage in recent works. To that effect, we propose to measure the data consistency quantitatively by some simplification and assumptions. Using the measurement, we analyze existing techniques for inverting samples and get some insightful information that inspires a novel loss function to reduce the inconsistency. Specifically, the loss minimizes the KL divergence of the distributions of inverted and real data under the tied multivariate Gaussian assumption, which is easy to implement in continual learning. In addition, we observe that the norms of old class weights turn to decrease continually as learning progresses. We thus analyze the underlying reasons and propose a simple regularization term to balance the class weights so that the samples of old classes are more distinguishable. To conclude, we propose the Consistency enhanced data replay with debiased classifier for Class Incremental Learning (CCIL). Extensive experiments on CIFAR-100, Tiny-ImageNet, and ImageNet100 show consistently improved performance of CCIL compared to previous approaches.

An Efficient Replay for Class-Incremental Learning with Pre-trained Models

Dynamic Replay Training for Class-Incremental Learning

Generative Feature Replay For Class-Incremental Learning

Memory-Free Generative Replay For Class-Incremental Learning

Squeezing More Past Knowledge for Online Class-Incremental Continual Learning

Enhancing Consistency and Mitigating Bias: A Data Replay Approach for Incremental Learning

A Continual Learning Method for Reducing Class Interference Based on Replay

Continual Recognition with Adaptive Memory Update.

Detachedly Learn a Classifier for Class-Incremental Learning

Class-Incremental Learning Using Generative Experience Replay Based on Time-aware Regularization

Class-Incremental Learning using Diffusion Model for Distillation and Replay

Class Incremental Learning with Important and Diverse Memory.

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

PCR: Proxy-based Contrastive Replay for Online Class-Incremental Continual Learning

Replay-enhanced Continual Reinforcement Learning

Class-Incremental Learning Based on Big Dataset Pre-Trained Models

Balanced Destruction-Reconstruction Dynamics for Memory-Replay Class Incremental Learning

Defying Imbalanced Forgetting in Class Incremental Learning

Condensed Prototype Replay for Class Incremental Learning

MixER: Mixup-Based Experience Replay for Online Class-Incremental Learning

Model Behavior Preserving for Class-Incremental Learning