Active Learning via Regression Beyond Realizability
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_ | 1866912412611903488 |
|---|---|
| author | Ganju, Atul Aiyer, Shashaank Sriraman, Ved Sridharan, Karthik |
| author_facet | Ganju, Atul Aiyer, Shashaank Sriraman, Ved Sridharan, Karthik |
| contents | We present a new active learning framework for multiclass classification based on surrogate risk minimization that operates beyond the standard realizability assumption. Existing surrogate-based active learning algorithms crucially rely on realizability$\unicode{x2014}$the assumption that the optimal surrogate predictor lies within the model class$\unicode{x2014}$limiting their applicability in practical, misspecified settings. In this work we show that under conditions significantly weaker than realizability, as long as the class of models considered is convex, one can still obtain a label and sample complexity comparable to prior work. Despite achieving similar rates, the algorithmic approaches from prior works can be shown to fail in non-realizable settings where our assumption is satisfied. Our epoch-based active learning algorithm departs from prior methods by fitting a model from the full class to the queried data in each epoch and returning an improper classifier obtained by aggregating these models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00316 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Active Learning via Regression Beyond Realizability Ganju, Atul Aiyer, Shashaank Sriraman, Ved Sridharan, Karthik Machine Learning Statistics Theory We present a new active learning framework for multiclass classification based on surrogate risk minimization that operates beyond the standard realizability assumption. Existing surrogate-based active learning algorithms crucially rely on realizability$\unicode{x2014}$the assumption that the optimal surrogate predictor lies within the model class$\unicode{x2014}$limiting their applicability in practical, misspecified settings. In this work we show that under conditions significantly weaker than realizability, as long as the class of models considered is convex, one can still obtain a label and sample complexity comparable to prior work. Despite achieving similar rates, the algorithmic approaches from prior works can be shown to fail in non-realizable settings where our assumption is satisfied. Our epoch-based active learning algorithm departs from prior methods by fitting a model from the full class to the queried data in each epoch and returning an improper classifier obtained by aggregating these models. |
| title | Active Learning via Regression Beyond Realizability |
| topic | Machine Learning Statistics Theory |
| url | https://arxiv.org/abs/2506.00316 |