Bayesian Learning-driven Prototypical Contrastive Loss for Class-Incremental Learning
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912301431390208 |
|---|---|
| author | Raichur, Nisha L. Heublein, Lucas Feigl, Tobias Rügamer, Alexander Mutschler, Christopher Ott, Felix |
| author_facet | Raichur, Nisha L. Heublein, Lucas Feigl, Tobias Rügamer, Alexander Mutschler, Christopher Ott, Felix |
| contents | The primary objective of methods in continual learning is to learn tasks in a sequential manner over time (sometimes from a stream of data), while mitigating the detrimental phenomenon of catastrophic forgetting. This paper proposes a method to learn an effective representation between previous and newly encountered class prototypes. We propose a prototypical network with a Bayesian learning-driven contrastive loss (BLCL), tailored specifically for class-incremental learning scenarios. We introduce a contrastive loss that incorporates novel classes into the latent representation by reducing intra-class and increasing inter-class distance. Our approach dynamically adapts the balance between the cross-entropy and contrastive loss functions with a Bayesian learning technique. Experimental results conducted on the CIFAR-10, CIFAR-100, and ImageNet100 datasets for image classification and images of a GNSS-based dataset for interference classification validate the efficacy of our method, showcasing its superiority over existing state-of-the-art approaches. Git: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/gnss_class_incremental_learning |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_11067 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Bayesian Learning-driven Prototypical Contrastive Loss for Class-Incremental Learning Raichur, Nisha L. Heublein, Lucas Feigl, Tobias Rügamer, Alexander Mutschler, Christopher Ott, Felix Computer Vision and Pattern Recognition Artificial Intelligence 62P30, 68T30, 68T05, 68T37 G.3; I.2.4; I.2.6 The primary objective of methods in continual learning is to learn tasks in a sequential manner over time (sometimes from a stream of data), while mitigating the detrimental phenomenon of catastrophic forgetting. This paper proposes a method to learn an effective representation between previous and newly encountered class prototypes. We propose a prototypical network with a Bayesian learning-driven contrastive loss (BLCL), tailored specifically for class-incremental learning scenarios. We introduce a contrastive loss that incorporates novel classes into the latent representation by reducing intra-class and increasing inter-class distance. Our approach dynamically adapts the balance between the cross-entropy and contrastive loss functions with a Bayesian learning technique. Experimental results conducted on the CIFAR-10, CIFAR-100, and ImageNet100 datasets for image classification and images of a GNSS-based dataset for interference classification validate the efficacy of our method, showcasing its superiority over existing state-of-the-art approaches. Git: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/gnss_class_incremental_learning |
| title | Bayesian Learning-driven Prototypical Contrastive Loss for Class-Incremental Learning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence 62P30, 68T30, 68T05, 68T37 G.3; I.2.4; I.2.6 |
| url | https://arxiv.org/abs/2405.11067 |