Multi-Task Active Learning with Efficient Resource Allocation
Abstract
Many scientific studies allow costly auxiliary information to be collected during data labeling but not at deployment. Examples include diagnostic tests, laboratory assays, and expert evaluations. We study prediction under this deployment asymmetry, where auxiliary variables are selectively acquired during labeling under a budget constraint but systematically unavailable at prediction time, creating a missing-by-design problem that couples data acquisition, surrogate construction, and prediction. In this work, we introduce Active Learning with Cost-Adaptive Task Resource Allocations (ALCATRAs), a unified framework for selectively acquiring auxiliary information under resource constraints and leveraging that information to improve downstream prediction. ALCATRAs consists of two main components: a task-selection policy which strategically selects a sequence of cost-effective tasks for unlabeled data to perform, and a surrogate learning procedure which transfers knowledge from completed tasks to enhance model predictions. In theory, we show the effectiveness of surrogate models and sample-efficient task policies in improving the model's prediction error bound. Simulation studies and an application to the UCI heart disease cohort demonstrate improved sample efficiency of the proposed ALCATRAs framework relative to baselines under the studied settings.
Related
- Efficient Joint Trajectory Optimization and Resource Allocation for Secure UAV Networks: A Deep Reinforcement Learning Approach
- Multi-view Semantic-aware Resource Allocation for Edge AI
- Strategic Multi-Agent Resource Allocation via Statistical Contracts
- A Token/KV-Cache Communication Media Selection and Resource Allocation Strategy for Multi-Agent Collaboration
- Efficient Active Auditing of Multi-Group Fairness with Bias Probes
BibTeX
@misc{ye2026multitask,
title = {{Multi-Task Active Learning with Efficient Resource Allocation}},
author = {Hanwen Ye and Jiuchen Zhang and Annie Qu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.07045},
eprint = {2610.07045},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.07045},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Multi-Task Active Learning with Efficient Resource Allocation
AU - Hanwen Ye
AU - Jiuchen Zhang
AU - Annie Qu
PY - 2026
DA - 2026/10/05
PB - arXiv
DO - 10.48550/arxiv.2610.07045
UR - https://arxiv.org/abs/2610.07045
AB - Many scientific studies allow costly auxiliary information to be collected during data labeling but not at deployment. Examples include diagnostic tests, laboratory assays, and expert evaluations. We study prediction under this deployment asymmetry, where auxiliary variables are selectively acquired during labeling under a budget constraint but systematically unavailable at prediction time, creating a missing-by-design problem that couples data acquisition, surrogate construction, and prediction. In this work, we introduce Active Learning with Cost-Adaptive Task Resource Allocations (ALCATRAs), a unified framework for selectively acquiring auxiliary information under resource constraints and leveraging that information to improve downstream prediction. ALCATRAs consists of two main components: a task-selection policy which strategically selects a sequence of cost-effective tasks for unlabeled data to perform, and a surrogate learning procedure which transfers knowledge from completed tasks to enhance model predictions. In theory, we show the effectiveness of surrogate models and sample-efficient task policies in improving the model's prediction error bound. Simulation studies and an application to the UCI heart disease cohort demonstrate improved sample efficiency of the proposed ALCATRAs framework relative to baselines under the studied settings.
ER -