Active Learning for Stochastic Contextual Linear Bandits
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916042613194752 |
|---|---|
| author | Brunskill, Emma Karmarkar, Ishani Li, Zhaoqi |
| author_facet | Brunskill, Emma Karmarkar, Ishani Li, Zhaoqi |
| contents | A key goal in stochastic contextual linear bandits is to efficiently learn a near-optimal policy. Prior algorithms for this problem learn a policy by strategically sampling actions but naively (passively) sampling contexts from the underlying context distribution. However, in many practical scenarios -- including online content recommendation, survey research, and clinical trials -- practitioners can actively sample or recruit contexts based on prior knowledge of the context distribution. Despite this potential for active learning, the role of strategic context sampling in stochastic contextual linear bandits is underexplored. We propose an algorithm that learns a near-optimal policy by strategically sampling rewards of context-action pairs. We prove instance-dependent theoretical guarantees demonstrating that our active context sampling strategy can improve over the minimax rate by up to a factor of $\sqrt{d}$, where $d$ is the linear dimension. We show empirically that our algorithm reduces the number of samples needed to learn a near-optimal policy, in tasks such as warfarin dose prediction and joke recommendation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_24803 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Active Learning for Stochastic Contextual Linear Bandits Brunskill, Emma Karmarkar, Ishani Li, Zhaoqi Machine Learning A key goal in stochastic contextual linear bandits is to efficiently learn a near-optimal policy. Prior algorithms for this problem learn a policy by strategically sampling actions but naively (passively) sampling contexts from the underlying context distribution. However, in many practical scenarios -- including online content recommendation, survey research, and clinical trials -- practitioners can actively sample or recruit contexts based on prior knowledge of the context distribution. Despite this potential for active learning, the role of strategic context sampling in stochastic contextual linear bandits is underexplored. We propose an algorithm that learns a near-optimal policy by strategically sampling rewards of context-action pairs. We prove instance-dependent theoretical guarantees demonstrating that our active context sampling strategy can improve over the minimax rate by up to a factor of $\sqrt{d}$, where $d$ is the linear dimension. We show empirically that our algorithm reduces the number of samples needed to learn a near-optimal policy, in tasks such as warfarin dose prediction and joke recommendation. |
| title | Active Learning for Stochastic Contextual Linear Bandits |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.24803 |