Few-Shot Incremental Learning with Continually Evolved Classifiers
Few-shot class-incremental learning (FSCIL) aims to design machine learning algorithms that can continually learn new concepts from a few data points, without forgetting knowledge of old classes. The difficulty lies in that limited data from new classes not only lead to significant overfitting issues but also exacerbate the notorious catastrophic forgetting problems. Moreover, as training data come in sequence in FSCIL, the learned classifier can only provide discriminative information in individual sessions, while FSCIL requires all classes to be involved for evaluation. In this paper, we address the FSCIL problem from two aspects. First, we adopt a simple but effective decoupled learning strategy of representations and classifiers that only the classifiers are updated in each incremental session, which avoids knowledge forgetting in the representations. By doing so, we demonstrate that a pre-trained backbone plus a non-parametric class mean classifier can beat state-of-the-art methods. Second, to make the classifiers learned on individual sessions applicable to all classes, we propose a Continually Evolved Classifier (CEC) that employs a graph model to propagate context information between classifiers for adaptation. To enable the learning of CEC, we design a pseudo incremental learning paradigm that episodically constructs a pseudo incremental learning task to optimize the graph parameters by sampling data from the base dataset. Experiments on three popular benchmark datasets, including CIFAR100, miniImageNet, and Caltech-USCD Birds-200-2011 (CUB200), show that our method significantly outperforms the baselines and sets new state-of-the-art results with remarkable advantages.
Code (1)
Tasks
class-incremental learningClass Incremental LearningFew-Shot Class-Incremental LearningIncremental LearningSimilar Papers 제목 키워드 기반
Embedding Space Allocation with Angle-Norm Joint Classifiers for Few-Shot Class-Incremental Learning
Few-shot class-incremental learning (FSCIL) aims to continually learn new classes from only a few samples without forgetting previous ones, requiring intelligent agents to adapt to dynamic environments. FSCIL combines th…
class-incremental learningClass Incremental LearningFew-Shot Class-Incremental LearningFew-Shot Learning+1Bias Mitigating Few-Shot Class-Incremental Learning
Few-shot class-incremental learning (FSCIL) aims at recognizing novel classes continually with limited novel class samples. A mainstream baseline for FSCIL is first to train the whole model in the base session, then free…
class-incremental learningClass Incremental LearningFew-Shot Class-Incremental LearningIncremental LearningDemystifying the Base and Novel Performances for Few-shot Class-incremental Learning
Few-shot class-incremental learning (FSCIL) has addressed challenging real-world scenarios where unseen novel classes continually arrive with few samples. In these scenarios, it is required to develop a model that recogn…
class-incremental learningClass Incremental LearningFew-Shot Class-Incremental LearningIncremental LearningFew-shot Class-incremental Audio Classification Using Stochastic Classifier
It is generally assumed that number of classes is fixed in current audio classification methods, and the model can recognize pregiven classes only. When new classes emerge, the model needs to be retrained with adequate s…
Audio ClassificationClassificationActive Class Selection for Few-Shot Class-Incremental Learning
For real-world applications, robots will need to continually learn in their environments through limited interactions with their users. Toward this, previous works in few-shot class incremental learning (FSCIL) and activ…
class-incremental learningClass Incremental LearningFew-Shot Class-Incremental LearningIncremental Learning+1