mlsanityData to 14 Oct 2026

Adaptive Active Learning for Regression via Reinforcement Learning

arXiv 2603.10435 ↗First posted 11 Mar 2026stat.ML cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL theory and sample complexity

Abstract (arXiv)

Active learning for regression reduces labeling costs by selecting the most informative samples. Improved Greedy Sampling is a prominent method that balances feature-space diversity and output-space uncertainty using a static, multiplicative rule. We propose Weighted improved Greedy Sampling (WiGS), which replaces this framework with a dynamic, additive criterion. We formulate weight selection as a reinforcement learning problem, enabling an agent to adapt the exploration-investigation balance throughout learning. Experiments on 18 benchmark datasets and a synthetic environment show WiGS outperforms iGS and other baseline methods in both accuracy and labeling efficiency, particularly in domains with irregular data density where the baseline's multiplicative rule ignores high-error samples in dense regions.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL theory and sample complexity, one of 15 directions in Deep reinforcement learning methods. Builds on 8 of its 22 references that are in the map. First posted on arXiv 11 Mar 2026.

Builds on

8 of 22 references.

References · 8

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  3. 2017Learning Active Learning from Datacited by 101
  4. 2017Learning how to Active Learn: A Deep Reinforcement Learning Approachcited by 79
  5. 2023Navigating the Pitfalls of Active Learning Evaluation: A Systematic Framework for Meaningful Performance Assessmentcited by 26
  6. 2017Active One-shot Learningcited by 21
  7. 2022Active Learning for Regression by Inverse Distance Weightingcited by 9
  8. 2020Pool-Based Unsupervised Active Learning for Regression Using Iterative Representativeness-Diversity Maximization (iRDM)cited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Fast and Robust: Task Sampling with Posterior and Diversity Synergies for Adaptive Decision-Makers in Randomized Environments
  2. 2024On Sample-Efficient Offline Reinforcement Learning: Data Diversity, Posterior Sampling, and Beyond
  3. 2025RL as Regressor: A Reinforcement Learning Approach for Function Approximation
  4. 2026Fine-Tuning on Self-Generated and Reward-Weighted Data: Learning Dynamics, Convergence Rates, and Benefits of Off-Policyness
  5. 2024Towards Sample-Efficiency and Generalization of Transfer and Inverse Reinforcement Learning: A Comprehensive Literature Review
  6. 2026Bayesian Symbolic Regression with Entropic Reinforcement Learning
  7. 2026Elucidating the Design Space of Regression-based Diffusion Reinforcement Learning
  8. 2022Environment Design for Inverse Reinforcement Learning