A New Benchmark for Few-Shot Class-Incremental Learning: Redefining the Upper Bound

2025-03-13 · Shiwon Kim, Dongjun Hwang, Sungwon Woo, Rita Singh

Class-incremental learning (CIL) aims to continuously adapt to emerging classes while retaining knowledge of previously learned ones. Few-shot class-incremental learning (FSCIL) presents an even greater challenge which requires the model to learn incremental classes with only a limited number of samples. In conventional CIL, joint training is widely considered the upper bound, serving as both a benchmark and a methodological guide. However, we find that joint training fails to be a meaningful upper bound in FSCIL due to the inherent difficulty of inter-task class separation (ICS) caused by severe class imbalance. In this work, we introduce a new joint training benchmark tailored for FSCIL by integrating imbalance-aware techniques, effectively bridging the performance gap between base and incremental classes. Furthermore, we point out inconsistencies in the experimental setup and evaluation of existing FSCIL methods. To ensure fair comparisons between different FSCIL approaches and joint training, we standardize training conditions and propose a unified evaluation protocol that simultaneously considers the validation set and computational complexity. By establishing a reliable upper bound and a standardized evaluation framework for FSCIL, our work provides a clear benchmark and a practical foundation for future research.

📄 PDF Abstract BibTeX arXiv:2503.10003

Code (0)

등록된 구현이 없습니다.

Tasks

class-incremental learning Class Incremental Learning Few-Shot Class-Incremental Learning Incremental Learning

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

BASE 설명 없음

Tell me what this is: Few-Shot Incremental Object Learning by a Robot

2020-07-15 · Ali Ayub, Alan R. Wagner

For many applications, robots will need to be incrementally trained to recognize the specific objects needed for an application. This paper presents a practical system for incrementally training a robot to recognize diff…

Incremental Learning

Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minima

2021-10-30 · NeurIPS 2021 12 · Guangyuan Shi, Jiaxin Chen, Wenlong Zhang, Li-Ming Zhan 외

This paper considers incremental few-shot learning, which requires a model to continually recognize new categories with only a few examples provided. Our study shows that existing methods severely suffer from catastrophi…

Few-Shot Class-Incremental Learning Few-Shot Learning Incremental Learning

Incremental Meta-Learning via Episodic Replay Distillation for Few-Shot Image Recognition

2021-11-09 · Kai Wang, Xialei Liu, Andy Bagdanov, Luis Herranz 외

Most meta-learning approaches assume the existence of a very large set of labeled data available for episodic meta-learning of base knowledge. This contrasts with the more realistic continual learning paradigm in which d…

Continual Learning Knowledge Distillation Meta-Learning

Cognitively-Inspired Model for Incremental Learning Using a Few Examples

2020-02-27 · Ali Ayub, Alan Wagner

Incremental learning attempts to develop a classifier which learns continuously from a stream of data segregated into different classes. Deep learning approaches suffer from catastrophic forgetting when learning classes …

class-incremental learning Class Incremental Learning General Classification Hippocampus+1

Few-shot Class-incremental Learning for Classification and Object Detection: A Survey

2023-08-13 · Jinghua Zhang, Li Liu, Olli Silvén, Matti Pietikäinen 외

Few-shot Class-Incremental Learning (FSCIL) presents a unique challenge in Machine Learning (ML), as it necessitates the Incremental Learning (IL) of new classes from sparsely labeled training samples without forgetting …

class-incremental learning Class Incremental Learning Class-Incremental Object Detection Few-Shot Class-Incremental Learning+5