Abstract:Class-Incremental Learning (CIL) aims to train a reliable model with the streaming data, which emerges unknown classes sequentially. Different from traditional closed set learning, CIL has two main challenges: (1) Novel class detection. The initial training data only contains incomplete classes, and streaming test data will accept unknown classes. Therefore, the model needs to not only accurately classify known classes, but also effectively detect unknown classes; (2) Model expansion. After the novel classes are detected, the model needs to be updated without re-training using the entire previous data. However, traditional CIL methods have not fully considered these two challenges. First, they are always restricted to single novel class detection within each phase and embedding confusion caused by unknown classes. Besides, they ignore the catastrophic forgetting of known categories in model update. To this end, we propose a semi-supervised style Class-Incremental Learning without Forgetting (CILF) method, which aims to learn adaptive embedding for processing novel class detection and model update in a unified framework. In detail, CILF designs to regularize classification with decoupled prototype based loss, which can improve the intra-class and inter-class structure significantly, and acquires a compact embedding representation for novel class detection in result. Then, CILF employs a learnable curriculum clustering operator to estimate the number of semantic clusters via fine-tuning the learned network, in which curriculum operator can adaptively learn the embedding in self-taught form. Therefore, CILF can detect multiple novel classes and mitigate the embedding confusion problem. Last, with the labeled streaming test data, CILF can update the network with robust regularization to mitigate the catastrophic forgetting. Consequently, CILF is able to iteratively perform novel class detection and model update. We verify the effectiveness of our model on four streaming classification tasks, and empirical studies show the superior performance of the proposed method.

Complementary Learning Subnetworks for Parameter-Efficient Class-Incremental Learning

A Robust and Anti-Forgettiable Model for Class-Incremental Learning

Efficient Non-Exemplar Class-Incremental Learning with Retrospective Feature Synthesis

Class Incremental Learning with Less Forgetting Direction and Equilibrium Point

Multi-granularity knowledge distillation and prototype consistency regularization for class-incremental learning

Harnessing Neural Unit Dynamics for Effective and Scalable Class-Incremental Learning

NeCa: Network Calibration for Class Incremental Learning.

Class Incremental Learning with Self-Supervised Pre-Training and Prototype Learning

Long-Tail Class Incremental Learning Via Independent Sub-prototype Construction

Learning Adaptive Embedding Considering Incremental Class

Non-Exemplar Class-Incremental Learning Via Adaptive Old Class Reconstruction

Class-Incremental Learning: A Survey

Connection-Based Knowledge Transfer for Class Incremental Learning

Semantic Knowledge Guided Class-Incremental Learning

Dense Network Expansion for Class Incremental Learning

Memory Efficient Class-Incremental Learning for Image Classification

Towards Non-Exemplar Semi-Supervised Class-Incremental Learning

Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learning

Adaptive Aggregation Networks for Class-Incremental Learning

Combining Past, Present and Future: A Self-Supervised Approach for Class Incremental Learning.