---
title: "On the Discrimination and Consistency for Exemplar-Free Class Incremental Learning"
authors:
  - "Tianqi Wang"
  - "Jingcai Guo"
  - "Depeng Li"
  - "Zhi Chen"
doi: "10.24963/ijcai.2025/715"
url: "https://ijcai.org/proceedings/2025/715"
pdf: "https://ijcai.org/proceedings/2025/0715.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "6424-6432"
track: "Main Track"
keywords:
  - "Machine Learning: ML: Cost-sensitive learning"
  - "Machine Learning: ML: Incremental learning"
  - "Computer Vision: CV: Transfer, low-shot, semi- and un- supervised learning"
---

# On the Discrimination and Consistency for Exemplar-Free Class Incremental Learning

Tianqi Wang, Jingcai Guo, Depeng Li, Zhi Chen

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 6424-6432. DOI: [10.24963/ijcai.2025/715](https://doi.org/10.24963/ijcai.2025/715)

## Abstract

Exemplar-free class incremental learning (EF-CIL) is a nontrivial task that requires continuously enriching model capability with new classes while maintaining previously learned knowledge without storing and replaying any old class exemplars. An emerging theory-guided framework for CIL trains task-specific models for a shared network, shifting the pressure of forgetting to task-id prediction. In EF-CIL, task-id prediction is more challenging due to the lack of inter-task interaction (e.g., replays of exemplars). To address this issue, we conduct a theoretical analysis of the importance and feasibility of preserving a discriminative and consistent feature space, upon which we propose a novel method termed DCNet. Concretely, it progressively maps class representations into a hyperspherical space, in which different classes are orthogonally distributed to achieve ample inter-class separation. Meanwhile, it also introduces compensatory training to adaptively adjust supervision intensity, thereby aligning the degree of intra-class aggregation. Extensive experiments and theoretical analysis verified the superiority of DCNet. Code is available at https://github.com/Tianqi-Wang1/DCNet.