Gradient Imbalance and solution in Online Continual learning
Most existing techniques for online continual learning are based on experience-replay. In this approach, a memory buffer is used to save some data from past tasks for dealing with catastrophic forgetting. In training, a small batch of data from the data stream of the current task and some sampled data from a memory buffer are used jointly to update or train the current model. In this paper, we study the experience replay-based approach from a new angle, gradient imbalance. We first investigate and analyze this phenomenon experimentally from two perspectives: imbalance of samples introduced by experience replay and sequence of classes introduced by incremental learning. To our knowledge, this problem has not been studied before and it significantly limits the performance of online continual learning. Based on observations from experiments and theoretical analysis, a new learning strategy and a new loss are proposed to deal with the problem. Empirical evaluation shows that GAD helps improve the online CL performance by more than 11% in accuracy.
Code (0)
등록된 구현이 없습니다.
Tasks
Continual LearningIncremental LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Online Analytic Exemplar-Free Continual Learning with Large Models for Imbalanced Autonomous Driving Task
In autonomous driving, even a meticulously trained model can encounter failures when facing unfamiliar scenarios. One of these scenarios can be formulated as an online continual learning (OCL) problem. That is, data come…
Autonomous DrivingContinual LearningExemplar-FreeAIR: Analytic Imbalance Rectifier for Continual Learning
Continual learning enables AI models to learn new data sequentially without retraining in real-world scenarios. Most existing methods assume the training data are balanced, aiming to reduce the catastrophic forgetting pr…
class-incremental learningClass Incremental LearningContinual LearningExemplar-Free+1PVBF: A Framework for Mitigating Parameter Variation Imbalance in Online Continual Learning
Online continual learning (OCL), which enables AI systems to adaptively learn from non-stationary data streams, is commonly achieved using experience replay (ER)-based methods that retain knowledge by replaying stored pa…
Continual LearningOnline Continual Learning via Multiple Deep Metric Learning and Uncertainty-guided Episodic Memory Replay -- 3rd Place Solution for ICCV 2021 Workshop SSLAD Track 3A Continual Object Classification
Online continual learning in the wild is a very difficult task in machine learning. Non-stationarity in online continual learning potentially brings about catastrophic forgetting in neural networks. Specifically, online …
Autonomous DrivingContinual LearningContrastive LearningMetric Learning+1Optimizing Class Distribution in Memory for Multi-Label Online Continual Learning
Online continual learning, especially when task identities and task boundaries are unavailable, is a challenging continual learning setting. One representative kind of methods for online continual learning is replay-base…
Continual Learning