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

by   Yuhang Zhou, et al.

Class incremental learning (CIL) aims to incrementally update a trained model with the new classes of samples (plasticity) while retaining previously learned ability (stability). To address the most challenging issue in this goal, i.e., catastrophic forgetting, the mainstream paradigm is memory-replay CIL, which consolidates old knowledge by replaying a small number of old classes of samples saved in the memory. Despite effectiveness, the inherent destruction-reconstruction dynamics in memory-replay CIL are an intrinsic limitation: if the old knowledge is severely destructed, it will be quite hard to reconstruct the lossless counterpart. Our theoretical analysis shows that the destruction of old knowledge can be effectively alleviated by balancing the contribution of samples from the current phase and those saved in the memory. Motivated by this theoretical finding, we propose a novel Balanced Destruction-Reconstruction module (BDR) for memory-replay CIL, which can achieve better knowledge reconstruction by reducing the degree of maximal destruction of old knowledge. Specifically, to achieve a better balance between old knowledge and new classes, the proposed BDR module takes into account two factors: the variance in training status across different classes and the quantity imbalance of samples from the current phase and memory. By dynamically manipulating the gradient during training based on these factors, BDR can effectively alleviate knowledge destruction and improve knowledge reconstruction. Extensive experiments on a range of CIL benchmarks have shown that as a lightweight plug-and-play module, BDR can significantly improve the performance of existing state-of-the-art methods with good generalization.


page 1

page 7

page 8

page 10


Self-Paced Imbalance Rectification for Class Incremental Learning

Exemplar-based class-incremental learning is to recognize new classes wh...

Incremental Prompting: Episodic Memory Prompt for Lifelong Event Detection

Lifelong event detection aims to incrementally update a model with new e...

IB-DRR: Incremental Learning with Information-Back Discrete Representation Replay

Incremental learning aims to enable machine learning models to continuou...

Memorizing Complementation Network for Few-Shot Class-Incremental Learning

Few-shot Class-Incremental Learning (FSCIL) aims at learning new concept...

Learning to Segment the Tail

Real-world visual recognition requires handling the extreme sample imbal...

DualMix: Unleashing the Potential of Data Augmentation for Online Class-Incremental Learning

Online Class-Incremental (OCI) learning has sparked new approaches to ex...

DiffusePast: Diffusion-based Generative Replay for Class Incremental Semantic Segmentation

The Class Incremental Semantic Segmentation (CISS) extends the tradition...

Please sign up or login with your details

Forgot password? Click here to reset