I study algorithms for artificial intelligence (AI) and decision making from the perspective of a theorist exploring fairness and human compatibility. As AI is increasingly deployed for real-world decision-making, it is crucial that we develop a principled theory and a robust practice of fairness in the algorithms we use. This agenda touches on a number of issues, motivated by societal applications, that demand increased sophistication of our algorithms in theory and practice. For examples:
- Rather than a single objective, we are increasingly realizing that algorithms for AI must handle multiple competing objectives, and we may wish to balance those in a nonlinear fashion.
- Algorithms should be interpretable where possible, and should otherwise always have explainable properties or guarantees, and should be auditable in principle.
- We must explicitly model uncertainty both about the world and about the objectives of our multiagent systems, to ensure safety and human
- In the age of big and sensitive data, we may wish to design systems that can guarantee rigorous notions of privacy such as are articulated in the theory of differential privacy.
My research agenda seeks to make progress toward a better artificial intelligence in these directions using techniques from machine learning, formal algorithm design, computational microeconomics, algorithmic game theory, and multiagent systems.
I am particularly committed to engaging undergraduate students in research in order to expand opportunities for access to scientific inquiry and graduate education. If you are a student at Duke interested in research, please do not hesitate to reach out to me in class or by email.
Working Papers
- Nianli Peng, Muhang Tian, and Brandon 2024. Multi-objective Reinforcement Learning with Nonlinear Preferences: Provable Approximation for Maximizing Expected Scalarized Return. Currently under review at the 38th Annual Conference on Neural Information Processing Systems (NeurIPS 2024).
- Kerry Lu and Brandon 2024. Proportionality and Free Riders: Committee Selection with Strategic Voters. Under review at the 20th Conference on Web and Internet Economics (WINE 2024).
Peer Reviewed Publications
Papers are listed in reverse chronological order by year, with ties broken alphabetically. Note that unlike many other disciplines in the sciences, most research in theory of algorithms and artificial intelligence is published in conference proceedings. These publications are competitively peer-reviewed.
- David Pujol, Albert Sun, Brandon Fain, and Ashwin Machanavajjhala. 2023. Multi-Analyst Differential Privacy for Online Query Answering. In Proceedings of the 49th International Conference on Very Large Data Bases (VLDB 2023).
- Zeyu Shen, Zhiyi Wang, Xingyu Zhu, Brandon Fain, and Kamesh Munagala. 2023. Fairness in the Assignment Problem with Uncertainty. In Proceedings of the 22nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2023).
- Zimeng Fan, Nianli Peng, Muhang Tian, and Brandon Fain. 2023. Welfare and Fairness in Multi-objective Reinforcement Learning. In Proceedings of the 22nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2023).
- Jerry Lin, Carolyn Chen, Marc Chmielewski, Samia Zaman, and Brandon Fain. 2022. Auditing for Gerrymandering by Identifying Disenfranchised Individuals. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT 2022).
- Liang Lyu, Brandon Fain, Kamesh Munagala, and Kangning Wang. 2021. Centrality with Diversity. In Proceedings of the 14th ACM International Conference on Web Search and Data Mining (WSDM 2021).
- David Pujol, Yikai Wu, Brandon Fain, and Ashwin Machanavajjhala. 2021. Budget Sharing for Multi-Analyst Differential Privacy. In Proceedings of the 47th International Conference on Very Large Data Bases (VLDB 2021).
- Brandon Fain, William Fan, and Kamesh Munagala. 2020. Concentration of Distortion: The Value of Extra Voters in Randomized Social Choice. In Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI 2020).
- Xingyu Chen, Brandon Fain, Liang Lyu, and Kamesh Munagala. 2019. Proportionally Fair Clustering. In Proceedings of the 36th International Conference on Machine Learning (ICML 2019).
- Brandon Fain, Ashish Goel, Kamesh Munagala, and Nina Prabhu. 2019. Random Dictators with a Random Referee: Constant Sample Complexity Mechanisms for Social Choice. In Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI 2019).
- Brandon Fain, Kamesh Munagala, and Nisarg Shah. 2018. Fair Allocation of Indivisible Public Goods. In Proceedings of the 2018 ACM Conference on Economics and Computation (EC 2018).
- Brandon Fain, Ashish Goel, Kamesh Munagala, and Sukolsak Sakshuwong. 2017. Sequential Deliberation for Social Choice. In Proceedings of the 13th International Conference on Web and Internet Economics (WINE 2017).
- Mayuresh Kunjir, Brandon Fain, Kamesh Munagala, and Shivnath Babu. 2017. ROBUS: Fair Cache Allocation for Data-parallel Workloads. In Proceedings of the 2017 ACM International Conference on Management of Data (SIGMOD 2017).
-
Brandon Fain, Ashish Goel, and Kamesh Munagala. 2016. The Core of The Participatory Budgeting Problem. In Proceedings of the 12th International Conference on Web and Internet Economics (WINE 2016).