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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Tianqi, Guo, Jingcai, Li, Depeng, Chen, Zhi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915122886213632
author Wang, Tianqi
Guo, Jingcai
Li, Depeng
Chen, Zhi
author_facet Wang, Tianqi
Guo, Jingcai
Li, Depeng
Chen, Zhi
contents 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 the proposed DCNet.
format Preprint
id arxiv_https___arxiv_org_abs_2501_15454
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Discrimination and Consistency for Exemplar-Free Class Incremental Learning
Wang, Tianqi
Guo, Jingcai
Li, Depeng
Chen, Zhi
Computer Vision and Pattern Recognition
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 the proposed DCNet.
title On the Discrimination and Consistency for Exemplar-Free Class Incremental Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.15454