MgSvF: Multi-Grained Slow versus Fast Framework for Few-Shot Class-Incremental Learning

By Hanbin Zhao, Yongjian Fu, Mintong Kang, Qi Tian, Fei Wu, Xi Li. 2020

151 citations recorded by the source corpus.

As a challenging problem, few-shot class-incremental learning (FSCIL) continually learns a sequence of tasks, confronting the dilemma between slow forgetting of old knowledge and fast adaptation to new knowledge. In this paper, we concentrate on this “slow versus fast” (SvF) dilemma to determine which knowledge components to be updated in a slow fashion or a fast fashion, and thereby balance old-knowledge preservation and new-knowledge adaptation. We propose a multi-grained SvF learning strategy to cope with the SvF dilemma from two different grains: intra-space (within the same feature space) and inter-space (between two different feature spaces). The proposed strategy designs a novel frequency-aware regularization to boost the intra-space SvF capability, and meanwhile develops a new feature space composition operation to enhance the inter-space SvF learning performance. With the multi-grained SvF learning strategy, our method outperforms the state-of-the-art approaches by a large margin.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Meta & Continual Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Han-Jia Ye
2. De-chuan Zhan
3. Da-Wei Zhou
4. Ziwei Liu
5. Hang Su
6. Jun Zhu
7. Qiwen Wang
8. Liyuan Wang

In Computer Vision and Pattern Recognition

1. Han-Jia Ye
2. De-chuan Zhan
3. Jun Zhu
4. Dacheng Tao
5. Da-Wei Zhou
6. Ziwei Liu
7. Hang Su
8. Luc Van Gool

In Meta & Continual Learning

1. Han-Jia Ye
2. De-chuan Zhan
3. Da-Wei Zhou
4. Liyuan Wang
5. Xingxing Zhang
6. Hang Su
7. Jun Zhu
8. Qiwen Wang

Citations in the corpus

• A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, background): “The extremely limited training samples exacerbate the overfitting of previously-learned representations to subsequent tasks, which can be alleviated by recent work such as preserving the topology of representations (TOPIC [404]), constructing an exemplar relation graph for knowledge distillation (ER”
• Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks (2022, background): “Few-shot class-incremental learning is recently proposed to tackle few-shot inputs in the class-incremental setting [66], [67].”
• Forward Compatible Few-Shot Class-Incremental Learning (2022, background): “Few-Shot Class-Incremental Learning: is recently proposed to address the few-shot inputs in the incremental learning scenario [1, 11, 24, 61].”
• When Video Classification Meets Incremental Classes (2021, background): “Recently, there has been a large body of research in class-incremental image classification [7, 61].”
• What and Where: Learn to Plug Adapters via NAS for Multidomain Learning (2020, background): “In life-long learning [38]–[42], LTG [43] proposed to expand the network architecture by NAS while retaining the previously learned knowledge.”
• Deep Class-Incremental Learning: A Survey (2023, background): “Backbone Expansion [20], [21], [22], [67], [68], [69], [70], [71]”
• Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learning (2024, background): “Class-Incremental Learning (CIL) : requires a learning system to continually absorb new class knowledge without forgetting existing ones [13, 14, 20, 22, 38, 57, 59, 74, 81, 82], which can be roughly divided into several categories.”
• Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning (2023, background): “Both CIL and FSCIL require a delicate balance between well adapting a model to novel classes and less forgetting of old classes (Zhao et al., 2021).”
• Class-Incremental Learning: A Survey (2023, unknown): “However, since the distributions of extracted features and low-fidelity images may differ from the raw images, an extra adaptation process is needed for these methods, adding to the algorithm’s complexity.”
• Memory-Efficient Class-Incremental Learning for Image Classification (2020, unknown): “incremental learning with deep models [33]–[46].”
• GKEAL: Gaussian Kernel Embedded Analytic Learning for Few-Shot Class Incremental Task (2023, methodology): “The FSCIL [2, 23, 32] jointly incorporates the settings of CIL and FSL by performing CIL tasks with each phase containing limited samples for new tasks.”