Information Templates: A New Paradigm for Intelligent Active Feature Acquisition
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910011390689280 |
|---|---|
| author | Huang, Hung-Tien Dinh, Dzung Oliva, Junier B. |
| author_facet | Huang, Hung-Tien Dinh, Dzung Oliva, Junier B. |
| contents | Active feature acquisition (AFA) is an instance-adaptive paradigm in which, at inference time, a policy sequentially chooses which features to acquire (at a cost) before predicting. Existing approaches either train reinforcement learning policies, which deal with a difficult MDP, or greedy policies that cannot account for the joint informativeness of features or require knowledge about the underlying data distribution. To overcome this, we propose Template-based AFA (TAFA), a non-greedy framework that learns a small library of feature templates -- sets of features that are jointly informative -- and uses this library of templates to guide the next feature acquisitions. Through identifying feature templates, the proposed framework not only significantly reduces the action space considered by the policy but also alleviates the need to estimate the underlying data distribution. Extensive experiments on synthetic and real-world datasets show that TAFA outperforms the existing state-of-the-art baselines while achieving lower overall acquisition cost and computation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_18380 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Information Templates: A New Paradigm for Intelligent Active Feature Acquisition Huang, Hung-Tien Dinh, Dzung Oliva, Junier B. Artificial Intelligence Active feature acquisition (AFA) is an instance-adaptive paradigm in which, at inference time, a policy sequentially chooses which features to acquire (at a cost) before predicting. Existing approaches either train reinforcement learning policies, which deal with a difficult MDP, or greedy policies that cannot account for the joint informativeness of features or require knowledge about the underlying data distribution. To overcome this, we propose Template-based AFA (TAFA), a non-greedy framework that learns a small library of feature templates -- sets of features that are jointly informative -- and uses this library of templates to guide the next feature acquisitions. Through identifying feature templates, the proposed framework not only significantly reduces the action space considered by the policy but also alleviates the need to estimate the underlying data distribution. Extensive experiments on synthetic and real-world datasets show that TAFA outperforms the existing state-of-the-art baselines while achieving lower overall acquisition cost and computation. |
| title | Information Templates: A New Paradigm for Intelligent Active Feature Acquisition |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2508.18380 |