About Me

“If I have seen further, it is by standing on the shoulders of giants.“
– Isaac Newton

My research philosophy is to work on problems that not only advance our understanding of the fundamental questions in optimization and machine learning, but often also has a positive impact on the society through ensuring algorithmic fairness.

To see the full list of publications, please visit Publications. Papers are numbered from ID1-ID30+ as in the list in Publications, and appear based on their area on this page.

To see some public repositories for code related to our research, please visit Code.

Algorithmic Fairness

In this thread, we highlight some papers that address fairness, accountability, transparency, ethics concerns with respect to various life-critical and impactful applications. Each paper is assigned a unique ID (ID1 – ID30+) based on Publications list.

1. [ID38] Mixed-Integer Projections for Automated Data Correction of EMRs Improve Predictions of Sepsis among Hospitalized Patients (arxiv)
   with Mehak Arora, Hassan Mortagy, Nathan Dwarshius, Andre L. Holder, Rishikesan Kamaleswaran
   Under submission, 2024.
2. [ID34]. Discovering Opportunities in New York City’s Discovery Program: an Analysis of Affirmative Action Mechanisms,
   with Yuri Faenza and Xuan Zhang. (arxiv)
   Extended abstract in EC 2023.
3. [ID37] Which Lp norm is the fairest? Approximations for fair facility location across all “p”.,
   with Mohit Singh and Jai Moondra.
   Extended abstract in EC 2023.
4. [ID36]. Using Algorithms to Tame Discrimination: A Path to Algorithmic Diversity, Equity, and Inclusion,
   with Deven Desai and Jad Salem.
   UC Davis Law Review, 2023. (ssrn)
5. [ID19]. Secretary Problems with Biased Evaluations using Partial Ordinal Information,
   with Jad Salem,
   Management Science, 2023.
   Preliminary version at Web and Internet Economics WINE 2020 (Titled: Closing the GAP: Mitigating Bias in Online Résumé-Filtering),
   Mechanism Design for Social Good Workshop MD4SG,
   media coverage by Diginomica and CoE News GT (ssrn)
6. [ID18.] Reducing the Filtering Effect in Public School Admissions: A Bias-aware Analysis for Targeted Interventions,
   with Yuri Faenza, Xuan Zhang,
   Major revision from M&SOM 2023,
   Extended Abstract at SIAM ACDA 2023 (arxiv)
7. [ID29]. Fair and Reliable Reconnections for Temporary Disruptions in Electric Distribution Networks using Submodularity,
   with Cyrus Hettle and Daniel Molzahn, 
   major revision at INFORMS Journal on Computing, 2023. (arxiv)
8. [ID35]. Mathematically Quantifying Gerrymandering and Non-Responsiveness of the 2021 Georgia Congressional Districting Plan,
   with Zhanzhan Zhao, Cyrus Hettle, Jonathan Mattingly, Dana Randall, and Greg Herschlag,
   ACM conference on Equity and Access in Algorithms, Mechanisms, and Optimization EAAMO’22, 2022. (arxiv)
9. [ID32]. Don’t let Ricci v. DeStefano Hold You Back: A Bias-Aware Legal Solution to the Hiring Paradox,
   with Jad Salem and Deven Desai,
   ACM Conference on Fairness, Accountability and Transparency, FAccT 2022. (arxiv)
10. [ID27]. Algorithmic Challenges in Ensuring Fairness at the Time of Decision,
    with Jad Salem and Vijay Kamble, 
    Major revision in Operations Research 2023.
    Extended abstract in WINE 2022. (arxiv)
    Revenue Management and Pricing Conference, 2021
    Manufacturing and Service Operations Management (MSOM) Conference, 2021.
11. [ID14]. Individual Fairness in Hindsight
    with Vijay Kamble,
    Journal of Machine Learning Research JMLR 2021.
    Extended abstract in EC 2019,
    Spotlight talk at NeurIPS Workshop on Ethical, Social and Governance Issues in AI 2018. (workshop link) (arxiv)
12. [ID26]. Balanced Districting on Grid Graphs with Provable Compactness and Contiguity,
    with Cyrus Hettle, Shixiang Zhu and Yao Xie,
    Extended Abstract in Foundations of Responsible Computing, FORC 2021 (non-archival). (arxiv)
13. [ID22]. Group-Fair Online Allocation in Continuous Time,
    with Semih Cayci, Atilla Eryilmaz,
    34th Conference on Neural Information and Processing Systems, NeurIPS 2020 (arxiv)
14. [ID16]. Too many fairness metrics: Is there a solution? Equity across Demographic Groups for the Facility Location Problem
    with Akhil Jalan, Gireeja Ranade, Helen Yang, Simon Zhuang,
    Ethics of Data Science Conference EDSC 2020 (postponed to 2021),
    Mechanism Design for Social Good Workshop MD4SG 2020.
    Conditionally accepted to the Fields Institute Communication Series 2023. (ssrn)
15. [ID12]. Fairness in Inventory Routing
    with M. Wang,
    NeurIPS Workshop on Ethical, Social & Governance Issues in AI 2018 (link)

Optimization and Machine Learning

In this thread, we highlight some work that exploits combinatorial structure for:
– approximation algorithms (Section A) for problems in energy (Section D), quantum computation (Section E), and, revenue management and supply chains (Section F),
– faster running times by bridging discrete and continuous algorithms (Section B),
– online and machine learning algorithms (Section C),
– classical-quantum hybrid methods (Section E).
Publications are organized by technique and key application. Each paper is assigned a unique ID based on Publications list (ID1 to ID30+).

A. Approximation Algorithms

1. [ID37]. Which Lp norm is the fairest? Approximations for fair facility location across all “p”,
   with Mohit Singh and Jai Moondra.
   Extended abstract in EC 2023.
2. [ID31]. Hardness and Approximation of Submodular Minimum Linear Ordering Problems,
   with Majid Farhadi, Shengding Sun, Prasad Tetali, Michael Wigal.
   Mathematical Programming, 2023 (arxiv)
3. [ID29]. Fair and Reliable Reconnections for Temporary Disruptions in Electric Distribution Networks using Submodularity,
   with Cyrus Hettle, Swati Gupta, Daniel Molzahn.
   Major revision at INFORMS Journal on Computing, 2023 (arxiv)
4. [ID26]. Balanced Redistricting for Faster Emergency Response under Imbalanced Historic Data (arxiv) with Cyrus Hettle, Shixiang Zhu and Yao Xie.
   Extended abstract in FORC 2021 (non-archival track). 
5. [ID25]. Generating Target Graph Couplings for QAOA from Native Quantum Hardware Couplings (arxiv)
   with Joel Rajakumar, Jai Moondra and Creston Herold.
   Physical Review A, 2022. 
6. [ID20]. Electrical Flows over Spanning Trees,
   with Ali Khodabhaksh, Hassan Mortagy, Evdokia Nikolova.
   Mathematical Programming (series B), 2020 (arxiv)
7. [ID17]. An Efficient Algorithm for Dynamic Pricing using a Graphical Representation
   with Maxime Cohen, Jeremy J Kalas and Georgia Perakis.
   Production and Operations Management POMS 2020, Huffington Post Article (ssrn | link)
8. [ID16]. Too many fairness metrics: Is there a solution? Equity across Demographic Groups for Facility Location Problem,
   with Akhil Jalan, Gireeja Ranade, Helen Yang, Simon Zhuang.
   Fields Institute Communication Series, 2023.
   Preliminary version accepted to appear at Ethics of Data Science Conference EDSC 2020 (postponed to 2021), Mechanism Design for Social Good Workshop MD4SG 2020 (ssrn)
9. [ID11]. A 4/3 approximation for TSP on cubic 3-edge connected graphs (link)
   with Nishita Agarwal, Naveen Garg
   Operations Research Letters, 2018. Older version posted January 2011 on arxiv | Poster
10. [ID2]. Towards a 4/3-approximation for the Metric Traveling Salesman Problem (PDF),
    Swati Gupta, Master’s Thesis, IIT Delhi 2011

B. Bridging Discrete and Continuous Optimization

1. [ID30]. Reusing Combinatorial Structure: Faster Iterative Projections over Submodular Base Polytopes, with Jai Moondra, Hassan Mortagy.
   35th Conference on Neural Information and Processing Systems, NeurIPS 2021. (arxiv)
2. [ID24]. Bridging Classical and Quantum using SDP initialized warm-starts for QAOA,
   with Reuben Tate, Majid Farhadi, Creston Herold, Greg Mohler.
   ACM Transactions on Quantum Computing, 2022 (arxiv)
3. [ID21]. Walking in the Shadow: A New Perspective on Descent Directions for Constrained Minimization, with Hassan Mortagy and Sebastian Pokutta.
   34th Conference on Neural Information and Processing Systems, NeurIPS 2020 (arxiv)
4. [ID20]. Electrical Flows over Spanning Trees,
   with Ali Khodabhaksh, Hassan Mortagy, Evdokia Nikolova
   Mathematical Programming (Series B), 2020 (arxiv)
5. [ID13]. Limited Memory Kelley’s Method Converges for Composite Convex and Submodular Objectives(arxiv) with Madeleine Udell, Song Zhou.
   Thirty Second Conference on Neural Information and Processing Systems, NeurIPS (spotlight) 2018. 
   ++ honorable mention in the INFORMS Undergraduate OR Prize 2018 
6. [ID7]. Solving Combinatorial Games using Products, Projections and Lexicographically Optimal Bases (arxiv, video) with Michel Goemans and Patrick Jaillet, LIDS/MIT article.
   Optimization for ML Workshop at NeurIPS, 2016 (link)
7. [ID6]. Newton’s Method for Parametric Submodular Function Minimization (link | PDF) with Michel Goemans and Patrick Jaillet
   Integer Programming and Combinatorial Optimization, IPCO 2017
8. [ID4]. Combinatorial Structure in Online and Convex Optimization (PDF),
   Swati Gupta, PhD Thesis, MIT 2017

C. Online and Machine Learning

1. [ID28]. Sequential Sampling for Functional Estimation via SIEVE,
   with Alessia Benevento, Massimo Pacella and Kamran Paynabar.
   Quality and Reliability Engineering, 2024.
2. [ID27]. Algorithmic Challenges in Ensuring Fairness at the Time of Decision,
   with Jad Salem and Vijay Kamble.(arxiv)
   Major Revision, Operations Research.
   Preliminary version in WINE 2022.
   Revenue Management and Pricing Conference, 2021
   Manufacturing and Service Operations Management (MSOM) Conference, 2021.
3. [ID22]. Group-Fair Online Allocation in Continuous Time (arxiv),
   with Semih Cayci, Atilla Eryilmaz,
   NeurIPS 2020. 
4. [ID15]. Robust Classifiers using Robust Feature Augmentation (arxiv)
   with Kevin Eykholt, Atul Prakash, Amir Rahmati, Pratik Vaishnavi, Haizhong Zheng, 2019
5. [ID14]. Individual Fairness in Hindsight (arxiv),
   with Vijay Kamble,
   Journal of Machine Learning Research JMLR 2021. 
   20th ACM Conference on Economics and Computation, EC 2019. 
   NeurIPS Workshop on Ethical, Social and Governance Issues in AI (2018). (workshop link) ++ Spotlight talk in the NIPS Workshop on Ethical, Social and Governance Issues in AI 2018
6. [ID10]. Computational Comparison of Metaheuristics (link)
   with John Silberholz, Bruce Golden, Xingyin Wang,
   Handbook of Metaheuristics, Springer, 2019
7. [ID8]. What works best when? A Framework for Systematic Heuristic Evaluation (link | github)
   with John Silberholz and Iain Dunning,
   INFORMS Journal on Computing, IJOC, 2018. ++ Received a Special Recognition in the INFORMS Computing Society Student Paper Competition 2016. Finalist in Michigan Institute for Data Science’s (MIDAS) Reproducibility Challenge 2020.

D. Energy

1. [ID29]. Fair and Reliable Reconnections for Temporary Disruptions in Electric Distribution Networks using Submodularity,
   with Cyrus Hettle, Daniel Molzahn, 
   major revision in INFORMS Journal on Computing, 2023 (arxiv)
2. [ID20]. Electrical Flows over Spanning Trees,
   with Ali Khodabhaksh, Hassan Mortagy, Evdokia Nikolova,
   Mathematical Programming (Series B), 2020 (arxiv)
3. [ID9]. Robust Look-ahead Three-phase Balancing of Uncertain Distribution Loads
   with Le Xie, Xinbo Geng,
   Hawaii International Conference on System Sciences, HICSS 52, 2019,
   ++5th place in Best Industry Studies paper 2019, Simons Institute Research Vignette (arxiv)

E. Quantum Computation

1. [ID33]. Classically-inspired Mixers for QAOA Beat Goemans-Williamson’s Max-Cut at Low Circuit Depths,
   with Reuben Tate, Jai Moondra, Bryan Garg, Greg Mohler,
   Quantum, 2023. (arxiv)
2. [ID24]. Generating Target Graph Couplings for QAOA from Native Quantum Hardware Couplings (arxiv)
   with Joel Rajakumar, Jai Moondra and Creston Herold,
   Physical Review A, 2022. 
3. [ID23]. Bridging Classical and Quantum using SDP initialized warm-starts for QAOA,
   with Reuben Tate, Majid Farhadi, Creston Herold, Greg Mohler,
   ACM Transactions on Quantum Computing, 2022 (arxiv)

F. Revenue Management and Supply Chains

1. [ID27]. Algorithmic Challenges in Ensuring Fairness at the Time of Decision,
   with Jad Salem and Vijay Kamble, WINE 2022,
   major revision at Operations Research. (arxiv)
   Revenue Management and Pricing Conference, 2021
   Manufacturing and Service Operations Management (MSOM) Conference, 2021
2. [ID22]. Group-Fair Online Allocation in Continuous Time (arxiv)
   with Semih Cayci, Atilla Eryilmaz,
   NeurIPS 2020. 
3. [ID18]. Reducing the Filtering Effect in Public School Admissions: A Bias-aware Analysis for Targeted Interventions (arxiv)
   with Yuri Faenza, Xuan Zhang,
   Extended abstract in SIAM ACDA 2023,
   Major revision, M&SOM.
4. [ID17]. An Efficient Algorithm for Dynamic Pricing using a Graphical Representation
   with Maxime Cohen, Jeremy J Kalas and Georgia Perakis,
   Production and Operations Management POMS 2020,
   finalist for INFORMS Service Science Section Student Paper Competition 2016 | Huffington Post Article | ssrn | link
5. [ID16]. Too many fairness metrics: Is there a solution? (ssrn)
   with Akhil Jalan, Gireeja Ranade, Helen Yang, Simon Zhuang,
   Ethics of Data Science Conference, EDSC 2020 (postponed to 2021),
   Mechanism Design for Social Good Workshop, MD4SG 2020.
6. [ID14]. Individual Fairness in Hindsight (arxiv)
   with Vijay Kamble,
   Journal of Machine Learning Research Journal of Machine Learning Research 2021. 
   Economics and Computation, ACM EC 2019. 
   NeurIPS Workshop on Ethical, Social and Governance Issues in AI (2018). (workshop link)
   ++ Spotlight talk in the NIPS Workshop on Ethical, Social and Governance Issues in AI 2018
7. [ID12]. Fairness in Inventory Routing (link)
   with M. Wang,
   NeurIPS Workshop on Ethical, Social & Governance Issues in AI 2018. 
8. [ID5]. A Scalable Robust and Adaptive Optimization Approach to Inventory Routing (optonline) with Joel Tay and Dimitris Bertsimas, 2017.