I am an Associate Professor at the MIT Sloan School of Management in OR & Statistics and a Class of 1947 Career Development Professor. I am a core faculty member of the Operations Research Center, and affiliated with LIDS and CCSE. I received a Ph.D. in Operations Research from MIT and was fortunate to be advised by Michel Goemans and Patrick Jaillet. I hold a joint Masters and B.Tech in Computer Science from IIT Delhi.  My Erdös number is 2, and I’m on the scientific advisory board of Intrare.

Research Interests: The classical foundations of optimization were built for a world that no longer exists: data representing the ground truth, fixed objectives to be optimized, and single solutions in a non-interactive computational environment. Modern systems operate on noisy, biased, and heterogeneous data; the objectives are contested across stakeholders and increasingly expressed through human interaction and language; the computation interacts with users at different times. This creates new challenges that I’m developing innovative solutions for, within the broader context of law, policy, and society. My work has informed or been deployed in settings spanning hiring platforms, public school admissions, organ allocation, electric utility systems, districting and emergency response, AI infrastructure, and quantum hardware benchmarking. The three broad programs we are currently focused on are:

(1) Contextual and Noisy Data. Data carries the imprint of how it was made — social dynamics, measurement error, historical bias — and those flaws propagate unevenly into the decisions built on it. We develop both ordinal and cardinal models to confront this. Ordinal methods act on rankings rather than scores, sidestepping the contested numerical weightings that make decisions legally vulnerable; cardinal methods correct biased data directly and trace how the remaining uncertainty shapes downstream outcomes. Together, they have yielded new insights into matching markets, faster detection of critical health conditions, and novel scholarship mechanisms for disadvantaged students. Selected publications:

  • Reducing the Filtering Effect in Public School Admissions: A Bias-aware Analysis for Targeted Interventions – with Faenza, Vuorinen, Zhang. M&SOM (minor rev).
  • Improving Clinical Decision Support through Interpretable Machine Learning and Error Correction in Electronic Health Records – with M. Arora, H. Mortagy, N. Dwarshius, J. Wang, P. Yang, A. Holder, R. Kamaleswaran, Journal of the American Medical Informatics Association (JAMIA) 2025.
  • Using Algorithms to Tame Discrimination: A path to Diversity, Equity and Inclusion – with Desai, Salem. UC Davis Law Review 2023.
  • Secretary Problems with Biased Evaluations using Partial Ordinal Information – with J. Salem. Management Science 2023.

(2) Multi-criteria considerations. Real-world systems must balance competing demands — fairness, efficiency, reliability, compliance with law and policy — and no single objective function captures them all. Rather than commit to one formulation, our work builds provably small sets of solutions that collectively cover any formulation in a given class (portfolios), turning an unresolvable modeling debate into an actionable menu. We also develop methods that account for repeated user interactions over time, as data evolves and the system learns its parameters. Here, our algorithms constrain the entire trajectory of iterates in online learning and stochastic optimization — not just the endpoint — to balance fairness against efficiency along the way. Selected publications:

  • Portfolios:
    • Balancing Notions of Equity: Trade-offs Between Fair Portfolio Sizes and Achievable Guarantees – with J. Moondra, M. Singh, SODA 2025 (with invitation to Transactions on Algorithms). Mathematical Programming, 2026. 
    • Navigating the Social Welfare Frontier: Portfolios for Multi-objective Reinforcement Learning, with C. Kim, J. Moondra, S. Verma, M. Pollack, L. Kong, M. Tambe, ICML 2025.
  • Multi-criteria Guarantees:
    • Fair and Reliable Reconnections for Temporary Disruptions in Electric Distribution Networks – with C. Hettle and D. Molzahn, INFORMS Journal on Computing 2025.
    • Balanced Districting on Grid Graphs with Provable Compactness and Contiguity – with C. Hettle, S. Zhu, Y. Xie, FORC 2021. 
  • Trajectories:
    • Algorithmic Challenges in Ensuring Fairness at the Time of Decision – with J. Salem, V. Kamble, WINE 2022, and Operations Research 2025. 
    • Individual Fairness in Hindsight – with V. Kamble, Journal of Machine Learning Research 2021. 

(3) Optimization under Heterogeneous Computation. Faster computation requires opening the black boxes that traditional optimization methods typically assume, and exploiting the structure of decisions across computational paradigms. Our work first bridges discrete and continuous optimization, through new ways of warm-starting and rounding fractional solutions and carrying structural information between subproblems in iterative methods. A second bridge joins optimization and AI: we bring learning inside optimization subroutines. The third, looking ahead, spans classical and quantum compute: methods that compose classical optimization subroutines within quantum computation. Selected publications:

  • Bridging Discrete and Continuous Optimization:
    • Improved Regret Guarantees for Online Mirror Descent using a Portfolio of Mirror Maps – with J. Moondra, M. Singh. Under submission (arxiv)
    • Electrical Flows over Spanning Trees with A. Khodabhaksh, H. Mortagy, E. Nikolova Mathematical Programming B, 2022
    • Newton’s Method for Parametric Submodular Function Minimization with M. Goemans, P. Jaillet, IPCO 2017
  • Bridging Optimization and AI/ML:
    • TACOS: Topology-Aware Collective Algorithm Synthesizer for Distributed Machine Learning with Won, Elavazhagan, Srinivasan, Krishna, MICRO 2024
    • Reusing Combinatorial Structure: Faster Iterative Projections over Submodular Base Polytopes with J. Moondra, H. Mortagy NeurIPS 2021.
  • Hybrid Classical-Quantum Optimization:
    • Promise of Graph Sparsification and Decomposition for Noise Reduction in QAOA: Analysis for Trapped-Ion Compilations – with J. Moondra, G. Mohler, P. Lotshaw. Quantum 2026.
    • Quantum Optimization: Potential, Challenges, and the Path Forward – with 40+ authors, Nature Reviews Physics 2024.
    • Classically-inspired Mixers for QAOA Beat Goemans-Williamson’s Max-Cut at Low Circuit Depths, with R. Tate, J. Moondra, B. Gard, G. Mohler Quantum 2023. 

You can further explore my publications here. I’m fortunate to have the opportunity to work with many fantastic students (Research Group). If you are interested in joining the group, feel free to reach out.

Awards and Honors: I received the NSF CAREER Award in 2023, the JP Morgan Early Career Faculty Recognition in 2021, the NSF CISE Research Initiation Initiative Award in 2019, Simons Berkeley Research Fellowship 2017-2018, and the Google Women in Engineering Award (India) in 2011. I also led the technical thrust of Ethical AI in the multi-institution NSF AI Institute on Advances in Optimization (ai4opt.org), from 2021-2023, and was the Georgia Tech PI for DARPA award on Optimization for Trapped Ion Qubits from 2020-2024. My research is also supported by funding from cross-disciplinary initiatives at MIT, such as SERC and MIT-MGB HEALS. My group has received recognition at various venues like NeurIPS spotlight, INFORMS Doing Good with OR 2022, INFORMS Undergraduate Operations Research 2018, INFORMS Computing Society 2016, and INFORMS Service Science Student Paper 2016.

I am an associate editor for Open Journal of Mathematical Optimization. I previously served as a co-editor for OPTIMA (newsletter of the Mathematical Optimization Society). I am (or have served) on the PC for IPCO 2026, IPCO 2024, NeurIPS 2023 (area chair), ACDA 2023, EAAMO 2022, FAccT 2022 (area chair), WINE 2021 and FORC 2021 (publications chair), and award committees for INFORMS competitions. I have also guest-edited a special issue on Data Science and Optimization for the Fields Communication Series.

Contact. first name and last initial at mit.edu, E62-582 (or a coffee shop near MIT)
Support Staff: Natalie Jean (njean91 at mit.edu)

Miscellaneous.
For more information about the latest activity in my research group, check out News!
For more information about my educational activities and courses, check out Teaching.
For media inquiries, reach out to Natalie Jean (njean91 at mit.edu).
For my artistic pursuits and exhibits, check out ProofbyPalette.
So proud of my husband, Tushar Krishna, who just won the ACM-SIGARCH Maurice Wilkes Award 2026!