Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2502.11927 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910832730832896 |
|---|---|
| author | Mitchell, Rupert Alliegro, Antonio Camoriano, Raffaello Carrión-Ojeda, Dustin Carta, Antonio Chalvatzaki, Georgia Churamani, Nikhil D'Eramo, Carlo Hamidi, Samin Hesse, Robin Hinder, Fabian Kamath, Roshni Ramanna Lomonaco, Vincenzo Paul, Subarnaduti Pistilli, Francesca Tuytelaars, Tinne van de Ven, Gido M Kersting, Kristian Schaub-Meyer, Simone Mundt, Martin |
| author_facet | Mitchell, Rupert Alliegro, Antonio Camoriano, Raffaello Carrión-Ojeda, Dustin Carta, Antonio Chalvatzaki, Georgia Churamani, Nikhil D'Eramo, Carlo Hamidi, Samin Hesse, Robin Hinder, Fabian Kamath, Roshni Ramanna Lomonaco, Vincenzo Paul, Subarnaduti Pistilli, Francesca Tuytelaars, Tinne van de Ven, Gido M Kersting, Kristian Schaub-Meyer, Simone Mundt, Martin |
| contents | Continual learning (CL) is the sub-field of machine learning concerned with accumulating knowledge in dynamic environments. So far, CL research has mainly focused on incremental classification tasks, where models learn to classify new categories while retaining knowledge of previously learned ones. Here, we argue that maintaining such a focus limits both theoretical development and practical applicability of CL methods. Through a detailed analysis of concrete examples - including multi-target classification, robotics with constrained output spaces, learning in continuous task domains, and higher-level concept memorization - we demonstrate how current CL approaches often fail when applied beyond standard classification. We identify three fundamental challenges: (C1) the nature of continuity in learning problems, (C2) the choice of appropriate spaces and metrics for measuring similarity, and (C3) the role of learning objectives beyond classification. For each challenge, we provide specific recommendations to help move the field forward, including formalizing temporal dynamics through distribution processes, developing principled approaches for continuous task spaces, and incorporating density estimation and generative objectives. In so doing, this position paper aims to broaden the scope of CL research while strengthening its theoretical foundations, making it more applicable to real-world problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_11927 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Continual Learning Should Move Beyond Incremental Classification Mitchell, Rupert Alliegro, Antonio Camoriano, Raffaello Carrión-Ojeda, Dustin Carta, Antonio Chalvatzaki, Georgia Churamani, Nikhil D'Eramo, Carlo Hamidi, Samin Hesse, Robin Hinder, Fabian Kamath, Roshni Ramanna Lomonaco, Vincenzo Paul, Subarnaduti Pistilli, Francesca Tuytelaars, Tinne van de Ven, Gido M Kersting, Kristian Schaub-Meyer, Simone Mundt, Martin Machine Learning Continual learning (CL) is the sub-field of machine learning concerned with accumulating knowledge in dynamic environments. So far, CL research has mainly focused on incremental classification tasks, where models learn to classify new categories while retaining knowledge of previously learned ones. Here, we argue that maintaining such a focus limits both theoretical development and practical applicability of CL methods. Through a detailed analysis of concrete examples - including multi-target classification, robotics with constrained output spaces, learning in continuous task domains, and higher-level concept memorization - we demonstrate how current CL approaches often fail when applied beyond standard classification. We identify three fundamental challenges: (C1) the nature of continuity in learning problems, (C2) the choice of appropriate spaces and metrics for measuring similarity, and (C3) the role of learning objectives beyond classification. For each challenge, we provide specific recommendations to help move the field forward, including formalizing temporal dynamics through distribution processes, developing principled approaches for continuous task spaces, and incorporating density estimation and generative objectives. In so doing, this position paper aims to broaden the scope of CL research while strengthening its theoretical foundations, making it more applicable to real-world problems. |
| title | Continual Learning Should Move Beyond Incremental Classification |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2502.11927 |