Navigating Underspecified Optimization Problems using Portfolios for Human-AI Augmented Decision-Making

Abstract

Optimization has long furnished us with precise solutions when objectives and problem parameters are clearly defined. Yet with growing human–AI interactive systems and decision-makers facing real-world challenges—from healthcare to disaster management to infrastructure planning—the stated problem is often ambiguous (e.g., human to agentic AI: “prioritize allocation of resources to those with higher need”). In these regimes, one should not aim for a single answer, resorting to a black-box formulation, but rather construct a portfolio of high-quality solutions that represent the space of potential problem formulations. This is a key step towards democratizing access to optimization and AI, and navigating the space of possible formulations efficiently. I will present recent work on approximation algorithms for combinatorial optimization problems, such as facility location (EC 2023) and scheduling (SODA 2025), and demonstrate how these ideas can be extended to multi-stakeholder reinforcement learning (ICML 2025). Each of these results constructs a portfolio to provably approximate a class of generalized p-means or ordered norms. Portfolios provide compact yet representative sets of solutions that capture competing objectives (e.g., efficiency v/s fairness) and trade-offs. I will further discuss ongoing work on integrating these ideas into optimization algorithms, such as online mirror descent, where learning over a portfolio of mirror maps can lead to improved regret guarantees. Together, these results illustrate how optimization can expand beyond single-solution paradigms to provide the theoretical scaffolding for AI systems—creating menus of possibilities that support robust, adaptive, and participatory decision-making. This talk is based on multiple joint works with Mohit Singh, Milind Tambe, Jai Moondra, Cheol Woo Kim, Shresth Verma, Madeleine Pollack, and Lingkai Wong.

Bio

Swati Gupta is an Associate Professor and the Class of 1947 Career Development Professor at the MIT Sloan School of Management in the Operations Research and Statistics Group. Prior to this, she held a Fouts Family Early Career Professorship as an Assistant Professor at the Stewart School of Industrial & Systems Engineering at Georgia Tech from 2018-2023, where she served as the lead of Ethical AI in the NSF AI Institute on Advances in Optimization from 2021-2023. She received a Ph.D. in Operations Research from MIT in 2017, following a joint Masters and B.Tech in Computer Science from IIT Delhi in 2011. Her research bridges optimization, machine learning, and algorithmic fairness, to design algorithms that are both theoretically rigorous and socially impactful, with applications in healthcare, hiring, energy, quantum computing, and beyond. Her work has been recognized by the 2023 NSF CAREER Award, INFORMS Doing Good with OR 2022 (finalist), the JP Morgan Early Career Faculty Recognition in 2021, INFORMS Computing Society 2016 (special recognition), and the INFORMS Service Science Section 2016 (finalist).