Abstract:Few-shot class-incremental learning (FSCIL) aims to continually learn new classes using a few samples while not forgetting the old classes. The key of this task is effective knowledge transfer from the base session to the incremental sessions. Despite the advance of existing FSCIL methods, the proposed knowledge transfer learning schemes are sub-optimal due to the insufficient optimization for the model's plasticity. To address this issue, we propose a Random Episode Sampling and Augmentation (RESA) strategy that relies on diverse pseudo incremental tasks as agents to achieve the knowledge transfer. Concretely, RESA mimics the real incremental setting and constructs pseudo incremental tasks globally and locally, where the global pseudo incremental tasks are designed to coincide with the learning objective of FSCIL and the local pseudo incremental tasks are designed to improve the model's plasticity, respectively. Furthermore, to make convincing incremental predictions, we introduce a complementary model with a squared Euclidean-distance classifier as the auxiliary module, which couples with the widely used cosine classifier to form our whole architecture. By such a way, equipped with model decoupling strategy, we can maintain the model's stability while enhancing the model's plasticity. Extensive quantitative and qualitative experiments on three popular FSCIL benchmark datasets demonstrate that our proposed method, named Knowledge Transfer-driven Relation Complementation Network (KT-RCNet), outperforms almost all prior methods. More precisely, the average accuracy of our proposed KT-RCNet outperforms the second-best method by a margin of 5.26%, 3.49%, and 2.25% on miniImageNet, CIFAR100, and CUB200, respectively. Our code is available at <a class="link-external link-https" href="https://github.com/YeZiLaiXi/KT-RCNet.git" rel="external noopener nofollow">this https URL</a>.

Few-Shot Incremental Event Detection Method Based on External Knowledge for Sample Preservation

Class-Incremental Few-Shot Event Detection

Few-shot Incremental Event Detection

Meta-Learning with Dynamic-Memory-Based Prototypical Network for Few-Shot Event Detection.

Adaptive Knowledge-Enhanced Bayesian Meta-Learning for Few-shot Event Detection

Incremental Event Detection Via Knowledge Consolidation Networks

Knowledge-Based Fine-Grained Classification for Few-Shot Learning.

Few-shot Event Detection: An Empirical Study and a Unified View

Towards Generalized and Incremental Few-Shot Object Detection

Incremental Detection of Remote Sensing Objects with Feature Pyramid and Knowledge Distillation

Few-Shot Incremental Learning for Label-to-Image Translation

Prompt Learning with Knowledge Memorizing Prototypes for Generalized Few-Shot Intent Detection

Self-Promoted Prototype Refinement for Few-Shot Class-Incremental Learning

Continual Few-shot Event Detection via Hierarchical Augmentation Networks

Incremental Few-Shot Object Detection for Robotics

Few-Shot Class-Incremental Learning Via Feature Space Composition

Class-Incremental Few-Shot Object Detection

Few-Shot Self Reminder to Overcome Catastrophic Forgetting

Task Consistent Prototype Learning for Incremental Few-shot Semantic Segmentation

Knowledge Transfer-Driven Few-Shot Class-Incremental Learning

Prototypes Sampling Mechanism for Class Incremental Learning