Incremental Learning with Repetition via Pseudo-Feature Projection
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929734155239424 |
|---|---|
| author | Tscheschner, Benedikt Veas, Eduardo Masana, Marc |
| author_facet | Tscheschner, Benedikt Veas, Eduardo Masana, Marc |
| contents | Incremental Learning scenarios do not always represent real-world inference use-cases, which tend to have less strict task boundaries, and exhibit repetition of common classes and concepts in their continual data stream. To better represent these use-cases, new scenarios with partial repetition and mixing of tasks are proposed, where the repetition patterns are innate to the scenario and unknown to the strategy. We investigate how exemplar-free incremental learning strategies are affected by data repetition, and we adapt a series of state-of-the-art approaches to analyse and fairly compare them under both settings. Further, we also propose a novel method (Horde), able to dynamically adjust an ensemble of self-reliant feature extractors, and align them by exploiting class repetition. Our proposed exemplar-free method achieves competitive results in the classic scenario without repetition, and state-of-the-art performance in the one with repetition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_19922 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Incremental Learning with Repetition via Pseudo-Feature Projection Tscheschner, Benedikt Veas, Eduardo Masana, Marc Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Incremental Learning scenarios do not always represent real-world inference use-cases, which tend to have less strict task boundaries, and exhibit repetition of common classes and concepts in their continual data stream. To better represent these use-cases, new scenarios with partial repetition and mixing of tasks are proposed, where the repetition patterns are innate to the scenario and unknown to the strategy. We investigate how exemplar-free incremental learning strategies are affected by data repetition, and we adapt a series of state-of-the-art approaches to analyse and fairly compare them under both settings. Further, we also propose a novel method (Horde), able to dynamically adjust an ensemble of self-reliant feature extractors, and align them by exploiting class repetition. Our proposed exemplar-free method achieves competitive results in the classic scenario without repetition, and state-of-the-art performance in the one with repetition. |
| title | Incremental Learning with Repetition via Pseudo-Feature Projection |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.19922 |