mlsanityData to 14 Oct 2026

Active Learning with Safety Constraints

arXiv 2206.11183 ↗First posted 22 Jun 2022cs.LG stat.MLcited by 16

Not on the map. Why?

Abstract (arXiv)

Active learning methods have shown great promise in reducing the number of samples necessary for learning. As automated learning systems are adopted into real-time, real-world decision-making pipelines, it is increasingly important that such algorithms are designed with safety in mind. In this work we investigate the complexity of learning the best safe decision in interactive environments. We reduce this problem to a constrained linear bandits problem, where our goal is to find the best arm satisfying certain (unknown) safety constraints. We propose an adaptive experimental design-based algorithm, which we show efficiently trades off between the difficulty of showing an arm is unsafe vs suboptimal. To our knowledge, our results are the first on best-arm identification in linear bandits with safety constraints. In practice, we demonstrate that this approach performs well on synthetic and real world datasets.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Builds on 24 of its 47 references that are in the map. First posted on arXiv 22 Jun 2022.

Builds on

24 of 47 references.

References · 24

  1. 2018A Reductions Approach to Fair Classificationcited by 587
  2. 2014On the Complexity of Best Arm Identification in Multi-Armed Bandit Modelscited by 286
  3. 2013lil' UCB : An Optimal Exploration Algorithm for Multi-Armed Banditscited by 149
  4. 2014Best-Arm Identification in Linear Banditscited by 100
  5. 2019Sequential Experimental Design for Transductive Linear Banditscited by 91
  6. 2019Linear Stochastic Bandits Under Safety Constraintscited by 90
  7. 2018Stagewise Safe Bayesian Optimization with Gaussian Processescited by 79
  8. 2019Randomized Exploration in Generalized Linear Banditscited by 70
  9. 2020Stochastic Bandits with Linear Constraintscited by 68
  10. 2020Gamification of Pure Exploration for Linear Banditscited by 63
  11. 2016Conservative Contextual Linear Banditscited by 57
  12. 2021High-Dimensional Experimental Design and Kernel Banditscited by 56
12 more
  1. 2018Training Well-Generalizing Classifiers for Fairness Metrics and Other Data-Dependent Constraintscited by 53
  2. 2022Reward-Free RL is No Harder Than Reward-Aware RL in Linear Markov Decision Processescited by 49
  3. 2020An Empirical Process Approach to the Union Bound: Practical Algorithms for Combinatorial and Linear Banditscited by 45
  4. 2017Fully adaptive algorithm for pure exploration in linear banditscited by 43
  5. 2019Safe Linear Thompson Sampling with Side Informationcited by 37
  6. 2017The Simulator: Understanding Adaptive Sampling in the Moderate-Confidence Regimecited by 31
  7. 2020Stage-wise Conservative Linear Banditscited by 28
  8. 2021Best Arm Identification with Safety Constraintscited by 25
  9. 2021Nearly Optimal Algorithms for Level Set Estimationcited by 17
  10. 2021Improved Algorithms for Agnostic Pool-based Active Classificationcited by 15
  11. 2020A New Perspective on Pool-Based Active Classification and False-Discovery Controlcited by 9
  12. 2021Selective Sampling for Online Best-arm Identificationcited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 6; 2024: 4; 2025: 3; 2026: 2.

16 papers.

Sort:

  1. 2023Learning Safety Constraints from Demonstrations with Unknown Rewardscited by 19
  2. 2022Safe Linear Bandits over Unknown Polytopescited by 14
  3. 2023Pure Exploration in Bandits with Linear Constraintscited by 10
  4. 2023Directional Optimism for Safe Linear Banditscited by 8
  5. 2023Multitask Learning with No Regret: from Improved Confidence Bounds to Active Learningcited by 4
  6. 2024SaVeR: Optimal Data Collection Strategy for Safe Policy Evaluation in Tabular MDPcited by 3
  7. 2024Testing the Feasibility of Linear Programs with Bandit Feedbackcited by 3
  8. 2023Fair Active Learning in Low-Data Regimescited by 2
  9. 2024Pure Exploration for Constrained Best Mixed Arm Identification with a Fixed Budgetcited by 2
  10. 2025Asymptotically Optimal Linear Best Feasible Arm Identification with Fixed Budgetcited by 1
  11. 2025Multi-Metric Adaptive Experimental Design Under a Fixed Budget with Validationcited by 1
  12. 2023Incentivized Collaboration in Active Learningcited by 0
4 more
  1. 2024Learning to Explore with Lagrangians for Bandits under Unknown Linear Constraintscited by 0
  2. 2025Constrained Linear Thompson Samplingcited by 0
  3. 2026Latent Order Banditscited by 0
  4. 2026Multi-Task Representation Learning for Conservative Linear Banditscited by 0