Rethinking Few-Shot Class-Incremental Learning with Open-Set Hypothesis in Hyperbolic Geometry

TL;DR

This paper proposes a new approach for Few-Shot Class-Incremental Learning using hyperbolic neural networks and open-set hypothesis, improving recognition performance and handling overfitting with novel modules and comprehensive evaluation.

Contribution

It introduces Hyperbolic Reciprocal Point Learning and hyperbolic metric learning modules to enhance FSCIL by better managing open-set recognition and overfitting.

Findings

Improved performance on three benchmark datasets.

Effective handling of open-set and close-set recognition.

Reduced overfitting in few-shot learning scenarios.

Abstract

Few-Shot Class-Incremental Learning (FSCIL) aims at incrementally learning novel classes from a few labeled samples by avoiding the overfitting and catastrophic forgetting simultaneously. The current protocol of FSCIL is built by mimicking the general class-incremental learning setting, while it is not totally appropriate due to the different data configuration, i.e., novel classes are all in the limited data regime. In this paper, we rethink the configuration of FSCIL with the open-set hypothesis by reserving the possibility in the first session for incoming categories. To assign better performances on both close-set and open-set recognition to the model, Hyperbolic Reciprocal Point Learning module (Hyper-RPL) is built on Reciprocal Point Learning (RPL) with hyperbolic neural networks. Besides, for learning novel categories from limited labeled data, we incorporate a hyperbolic metric learning (Hyper-Metric) module into the distillation-based framework to alleviate the overfitting issue and better handle the trade-off issue between the preservation of old knowledge and the acquisition of new knowledge. The comprehensive assessments of the proposed configuration and modules on three benchmark datasets are executed to validate the effectiveness concerning three evaluation indicators.