# Learning to Decide with AI Assistance under Human-Alignment

arXiv: https://arxiv.org/abs/2605.12646 · HTML: https://mlsanity.com/paper/2605.12646/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 51 references that are in the map.
- First posted on arXiv 12 May 2026.

## Builds on (most cited first, 30 of 30)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Introduction to Multi-Armed Bandits](https://mlsanity.com/paper/1904.07272/) (2019)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making](https://mlsanity.com/paper/2001.02114/) (2020)
- [Towards a Science of Human-AI Decision Making: A Survey of Empirical Studies](https://mlsanity.com/paper/2112.11471/) (2021)
- [Being Optimistic to Be Conservative: Quickly Learning a CVaR Policy](https://mlsanity.com/paper/1911.01546/) (2019)
- [Combinatorial Multi-Armed Bandit with General Reward Functions](https://mlsanity.com/paper/1610.06603/) (2016)
- [Improving Expert Predictions with Conformal Prediction](https://mlsanity.com/paper/2201.12006/) (2022)
- [How model accuracy and explanation fidelity influence user trust](https://mlsanity.com/paper/1907.12652/) (2019)
- [Distribution-free calibration guarantees for histogram binning without sample splitting](https://mlsanity.com/paper/2105.04656/) (2021)
- [Oracle Efficient Online Multicalibration and Omniprediction](https://mlsanity.com/paper/2307.08999/) (2023)
- [Do Humans Trust Advice More if it Comes from AI? An Analysis of Human-AI Interactions](https://mlsanity.com/paper/2107.07015/) (2021)
- [The Role of Domain Expertise in User Trust and the Impact of First Impressions with Intelligent Systems](https://mlsanity.com/paper/2008.09100/) (2020)
- [Human-Aligned Calibration for AI-Assisted Decision Making](https://mlsanity.com/paper/2306.00074/) (2023)
- [Sequential estimation of quantiles with applications to A/B-testing and best-arm identification](https://mlsanity.com/paper/1906.09712/) (2019)
- [Optimal Thompson Sampling strategies for support-aware CVaR bandits](https://mlsanity.com/paper/2012.05754/) (2020)
- [Uncalibrated Models Can Improve Human-AI Collaboration](https://mlsanity.com/paper/2202.05983/) (2022)
- [Designing Decision Support Systems Using Counterfactual Prediction Sets](https://mlsanity.com/paper/2306.03928/) (2023)
- [Misplaced Trust: Measuring the Interference of Machine Learning in Human Decision-Making](https://mlsanity.com/paper/2005.10960/) (2020)
- [Towards Human-AI Complementarity with Prediction Sets](https://mlsanity.com/paper/2405.17544/) (2024)
- [From Calibration to Collaboration: LLM Uncertainty Quantification Should Be More Human-Centered](https://mlsanity.com/paper/2506.07461/) (2025)
- [Learning When to Advise Human Decision Makers](https://mlsanity.com/paper/2209.13578/) (2022)
- [Improving Screening Processes via Calibrated Subset Selection](https://mlsanity.com/paper/2202.01147/) (2022)
- [Towards Optimizing Human-Centric Objectives in AI-Assisted Decision-Making With Offline Reinforcement Learning](https://mlsanity.com/paper/2403.05911/) (2024)
- [Online Platt Scaling with Calibeating](https://mlsanity.com/paper/2305.00070/) (2023)
- [Useful Confidence Measures: Beyond the Max Score](https://mlsanity.com/paper/2210.14070/) (2022)
- [AI-Assisted Decision Making with Human Learning](https://mlsanity.com/paper/2502.13062/) (2025)
- [Human-Alignment Influences the Utility of AI-assisted Decision Making](https://mlsanity.com/paper/2501.14035/) (2025)
- [Human Misperception of Generative-AI Alignment: A Laboratory Experiment](https://mlsanity.com/paper/2502.14708/) (2025)
- [Narrowing Action Choices with AI Improves Human Sequential Decisions](https://mlsanity.com/paper/2510.16097/) (2025)

## Built on by (most cited first, 1 of 1)

- [Human-AI Teaming Through the Lens of Calibration](https://mlsanity.com/paper/2606.10906/) (2026)

## Abstract (arXiv)

It is widely agreed that when AI models assist decision-makers in high-stakes domains by predicting an outcome of interest, they should communicate the confidence of their predictions. However, empirical evidence suggests that decision-makers often struggle to determine when to trust a prediction based solely on this communicated confidence. In this context, recent theoretical and empirical work suggests a positive correlation between the utility of AI-assisted decision-making and the degree of alignment between the AI confidence and the decision-makers' confidence in their own predictions. Crucially, these findings do not yet elucidate the extent to which this alignment influences the complexity of learning to make optimal decisions through repeated interactions. In this paper, we address this question in the canonical case of binary predictions and binary decisions. We first show that this problem is equivalent to a two-armed online contextual learning problem with full feedback, and establish a lower bound of Omega (√|H| · |B| · T ) on the expected regret any learner can attain, where H and B denote the sets of human and AI confidence values. We then demonstrate that, under perfect alignment between AI and human confidence, a learner can attain an expected regret of O(√|H| · Tlog T) and, when √|H| = O(log T) and B is countable, a non-trivial generalization of the Dvoretzky-Kiefer-Wolfowitz inequality improves the regret bound to O(√Tlog T). Taken together, these results reveal that alignment can reduce the complexity of learning to make decisions with AI assistance. Experiments on real data from two different human-subject studies where participants solve simple decision-making tasks assisted by AI models show that our theoretical results are robust to violations of perfect alignment.