# Designing Algorithmic Recommendations to Achieve Human-AI Complementarity

arXiv: https://arxiv.org/abs/2405.01484 · HTML: https://mlsanity.com/paper/2405.01484/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 49 references that are in the map.
- First posted on arXiv 2 May 2024.

## Builds on (most cited first, 14 of 14)

- [Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practices](https://mlsanity.com/paper/1906.09208/) (2019)
- [From Predictive to Prescriptive Analytics](https://mlsanity.com/paper/1402.5481/) (2014)
- [The Algorithmic Automation Problem: Prediction, Triage, and Human Effort](https://mlsanity.com/paper/1903.12220/) (2019)
- [Towards a Science of Human-AI Decision Making: A Survey of Empirical Studies](https://mlsanity.com/paper/2112.11471/) (2021)
- [Human-Algorithm Collaboration: Achieving Complementarity and Avoiding Unfairness](https://mlsanity.com/paper/2202.08821/) (2022)
- [Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment](https://mlsanity.com/paper/2012.02845/) (2020)
- [Learning When to Advise Human Decision Makers](https://mlsanity.com/paper/2209.13578/) (2022)
- [Algorithmic Assistance with Recommendation-Dependent Preferences](https://mlsanity.com/paper/2208.07626/) (2022)
- [Safe Policy Learning through Extrapolation: Application to Pre-trial Risk Assessment](https://mlsanity.com/paper/2109.11679/) (2021)
- [The Rational Agent Benchmark for Data Visualization](https://mlsanity.com/paper/2304.03432/) (2023)
- [Policy Optimization for Personalized Interventions in Behavioral Health](https://mlsanity.com/paper/2303.12206/) (2023)
- [Ambiguous Dynamic Treatment Regimes: A Reinforcement Learning Approach](https://mlsanity.com/paper/2112.04571/) (2021)
- [Identifying Effects of Multivalued Treatments](https://mlsanity.com/paper/1805.00057/) (2018)
- [Probabilistic Prediction for Binary Treatment Choice: with focus on personalized medicine](https://mlsanity.com/paper/2110.00864/) (2021)

## Built on by (most cited first, 4 of 4)

- [AI Agents for Inventory Control: Human-LLM-OR Complementarity](https://mlsanity.com/paper/2602.12631/) (2026)
- [Human-AI Productivity Paradoxes: Modeling the Interplay of Skill, Effort, and AI Assistance](https://mlsanity.com/paper/2605.11350/) (2026)
- [2-Step Agent: A Framework for the Interaction of a Decision Maker with AI Decision Support](https://mlsanity.com/paper/2602.21889/) (2026)
- [Algorithmic Feature Highlighting for Human-AI Decision-Making](https://mlsanity.com/paper/2604.22236/) (2026)

## Abstract (arXiv)

Algorithms frequently assist, rather than replace, human decision-makers. However, the design and analysis of algorithms often focus on predicting outcomes and do not explicitly model their effect on human decisions. This discrepancy between the design and role of algorithmic assistants becomes particularly concerning in light of empirical evidence that suggests that algorithmic assistants again and again fail to improve human decisions. In this article, we formalize the design of recommendation algorithms that assist human decision-makers without making restrictive ex-ante assumptions about how recommendations affect decisions. We formulate an algorithmic-design problem that leverages the potential-outcomes framework from causal inference to model the effect of recommendations on a human decision-maker's binary treatment choice. Within this model, we introduce a monotonicity assumption that leads to an intuitive classification of human responses to the algorithm. Under this assumption, we can express the human's response to algorithmic recommendations in terms of their compliance with the algorithm and the active decision they would take if the algorithm sends no recommendation. We showcase the utility of our framework using an online experiment that simulates a hiring task. We argue that our approach can make sense of the relative performance of different recommendation algorithms in the experiment and can help design solutions that realize human-AI complementarity. Finally, we leverage our approach to derive minimax optimal recommendation algorithms that can be implemented with machine learning using limited training data.