Regularization Shortcomings for Continual Learning
In most machine learning algorithms, training data is assumed to be independent and identically distributed (iid). When it is not the case, the algorithm's performances are challenged, leading to the famous phenomenon of catastrophic forgetting. Algorithms dealing with it are gathered in the Continual Learning research field. In this paper, we study the regularization based approaches to continual learning and show that those approaches can not learn to discriminate classes from different tasks in an elemental continual benchmark: the class-incremental scenario. We make theoretical reasoning to prove this shortcoming and illustrate it with examples and experiments. Moreover, we show that it can have some important consequences on continual multi-tasks reinforcement learning or in pre-trained models used for continual learning. We believe that highlighting and understanding the shortcomings of regularization strategies will help us to use them more efficiently.
Code (0)
등록된 구현이 없습니다.
Tasks
Continual LearningMulti-Task LearningSimilar Papers 제목 키워드 기반
Bayesian Structure Adaptation for Continual Learning
Continual Learning is a learning paradigm where learning systems are trained with sequential or streaming tasks. Two notable directions among the recent advances in continual learning with neural networks are ($i$) varia…
Continual LearningA Unified Bayesian Framework for Discriminative and Generative Continual Learning
Continual Learning is a learning paradigm where learning systems are trained on a sequence of tasks. The goal here is to perform well on the current task without suffering from a performance drop on the previous tasks. T…
Continual LearningTriRE: A Multi-Mechanism Learning Paradigm for Continual Knowledge Retention and Promotion
Continual learning (CL) has remained a persistent challenge for deep neural networks due to catastrophic forgetting (CF) of previously learned tasks. Several techniques such as weight regularization, experience rehearsal…
Continual LearningDataset Condensation with Latent Quantile Matching
Dataset condensation (DC) methods aim to learn a smaller synthesized dataset with informative data records to accelerate the training of machine learning models. Current distribution matching (DM) based DC methods learn …
Dataset CondensationGraph LearningA Closer Look at Rehearsal-Free Continual Learning
Continual learning is a setting where machine learning models learn novel concepts from continuously shifting training data, while simultaneously avoiding degradation of knowledge on previously seen classes which may dis…
Continual LearningKnowledge DistillationL2 RegularizationNovel Concepts