Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2404.07729 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910406151241728 |
|---|---|
| author | Nasri, Nadia Gutiérrez-Álvarez, Carlos Lafuente-Arroyo, Sergio Maldonado-Bascón, Saturnino López-Sastre, Roberto J. |
| author_facet | Nasri, Nadia Gutiérrez-Álvarez, Carlos Lafuente-Arroyo, Sergio Maldonado-Bascón, Saturnino López-Sastre, Roberto J. |
| contents | Continual learning (CL) is crucial for evaluating adaptability in learning solutions to retain knowledge. Our research addresses the challenge of catastrophic forgetting, where models lose proficiency in previously learned tasks as they acquire new ones. While numerous solutions have been proposed, existing experimental setups often rely on idealized class-incremental learning scenarios. We introduce Realistic Continual Learning (RealCL), a novel CL paradigm where class distributions across tasks are random, departing from structured setups.
We also present CLARE (Continual Learning Approach with pRE-trained models for RealCL scenarios), a pre-trained model-based solution designed to integrate new knowledge while preserving past learning. Our contributions include pioneering RealCL as a generalization of traditional CL setups, proposing CLARE as an adaptable approach for RealCL tasks, and conducting extensive experiments demonstrating its effectiveness across various RealCL scenarios. Notably, CLARE outperforms existing models on RealCL benchmarks, highlighting its versatility and robustness in unpredictable learning environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_07729 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Realistic Continual Learning Approach using Pre-trained Models Nasri, Nadia Gutiérrez-Álvarez, Carlos Lafuente-Arroyo, Sergio Maldonado-Bascón, Saturnino López-Sastre, Roberto J. Machine Learning Computer Vision and Pattern Recognition Continual learning (CL) is crucial for evaluating adaptability in learning solutions to retain knowledge. Our research addresses the challenge of catastrophic forgetting, where models lose proficiency in previously learned tasks as they acquire new ones. While numerous solutions have been proposed, existing experimental setups often rely on idealized class-incremental learning scenarios. We introduce Realistic Continual Learning (RealCL), a novel CL paradigm where class distributions across tasks are random, departing from structured setups. We also present CLARE (Continual Learning Approach with pRE-trained models for RealCL scenarios), a pre-trained model-based solution designed to integrate new knowledge while preserving past learning. Our contributions include pioneering RealCL as a generalization of traditional CL setups, proposing CLARE as an adaptable approach for RealCL tasks, and conducting extensive experiments demonstrating its effectiveness across various RealCL scenarios. Notably, CLARE outperforms existing models on RealCL benchmarks, highlighting its versatility and robustness in unpredictable learning environments. |
| title | Realistic Continual Learning Approach using Pre-trained Models |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.07729 |