Balanced Gradient Sample Retrieval for Enhanced Knowledge Retention in Proxy-based Continual 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_ | 1866929639914471424 |
|---|---|
| author | Xu, Hongye Wasilewski, Jan Krawczyk, Bartosz |
| author_facet | Xu, Hongye Wasilewski, Jan Krawczyk, Bartosz |
| contents | Continual learning in deep neural networks often suffers from catastrophic forgetting, where representations for previous tasks are overwritten during subsequent training. We propose a novel sample retrieval strategy from the memory buffer that leverages both gradient-conflicting and gradient-aligned samples to effectively retain knowledge about past tasks within a supervised contrastive learning framework. Gradient-conflicting samples are selected for their potential to reduce interference by re-aligning gradients, thereby preserving past task knowledge. Meanwhile, gradient-aligned samples are incorporated to reinforce stable, shared representations across tasks. By balancing gradient correction from conflicting samples with alignment reinforcement from aligned ones, our approach increases the diversity among retrieved instances and achieves superior alignment in parameter space, significantly enhancing knowledge retention and mitigating proxy drift. Empirical results demonstrate that using both sample types outperforms methods relying solely on one sample type or random retrieval. Experiments on popular continual learning benchmarks in computer vision validate our method's state-of-the-art performance in mitigating forgetting while maintaining competitive accuracy on new tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_14430 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Balanced Gradient Sample Retrieval for Enhanced Knowledge Retention in Proxy-based Continual Learning Xu, Hongye Wasilewski, Jan Krawczyk, Bartosz Machine Learning Continual learning in deep neural networks often suffers from catastrophic forgetting, where representations for previous tasks are overwritten during subsequent training. We propose a novel sample retrieval strategy from the memory buffer that leverages both gradient-conflicting and gradient-aligned samples to effectively retain knowledge about past tasks within a supervised contrastive learning framework. Gradient-conflicting samples are selected for their potential to reduce interference by re-aligning gradients, thereby preserving past task knowledge. Meanwhile, gradient-aligned samples are incorporated to reinforce stable, shared representations across tasks. By balancing gradient correction from conflicting samples with alignment reinforcement from aligned ones, our approach increases the diversity among retrieved instances and achieves superior alignment in parameter space, significantly enhancing knowledge retention and mitigating proxy drift. Empirical results demonstrate that using both sample types outperforms methods relying solely on one sample type or random retrieval. Experiments on popular continual learning benchmarks in computer vision validate our method's state-of-the-art performance in mitigating forgetting while maintaining competitive accuracy on new tasks. |
| title | Balanced Gradient Sample Retrieval for Enhanced Knowledge Retention in Proxy-based Continual Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2412.14430 |