H. Milton Stewart School of Industrial and Systems Engineering, Georgia Tech.
Machine Learning Center (ML@GT),
Algorithms, Combinatorics and Optimization (ACO),
Algorithms, Randomness and Complexity Center (ARC).
Contact: email is <firstname> “g” at gatech dot edu
Bio: Assistant Professor, Georgia Tech (current). Research Fellow, Simons Institute (Aug 2017 – May 2018). Ph.D. in Operations Research, MIT 2017. Bachelors & Masters in Computer Science, IIT, Delhi 2011. (more)
Research Interests: Optimization. Combinatorial. Convex. Robust. Online and Machine Learning. Unintended Consequences. Bias and Fairness. Real-world Applications. (more)
Growing list of collaborators: Dimitris Bertsimas, Maxime Cohen, Iain Dunning, Naveen Garg, Michel Goemans, Bruce Golden, Robert Hildebrand, Patrick Jaillet, Vijay Kamble, Tushar Krishna, Kristen LeFevre, Shima Nassiri, Evdokia Nikolova, Georgia Perakis, Sebastian Pokutta, Atul Prakash, Gireeja Ranade, John Silberholz, Joel Tay, Amitabha Tripathi, Madeleine Udell, Xingyin Wang, Le Xie | IBM Research Zurich, Oracle Retail Data Science Group, Future of Privacy Forum.
01/2019: I am invited to speak at a panel on “Profiling, micro-targeting and a right to reasonable algorithmic inferences” organized by Microsoft at the International Conference on Computers, Privacy and Data Protection, in Brussels, January 30 – Feb 1, 2019.
01/2019: I will be a long-term visitor during the program on Fairness at the Simons Institute in Berkeley from May 30th to July 10th, 2019.
12/2018: Simons Reunion Workshop for Bridging Continuous and Discrete Optimization Program: presented joint work with Madeleine Udell and Sam Zhou on Limited Memory Kelley’s Method Converges for Composite Convex and Submodular Functions.
11/2018: Our workshop papers on Temporal Aspects of Individual Fairness (with Vijay Kamble) and Fairness in the Face of Uncertainty (with Michael Wang) got accepted in the NIPS Workshop on Ethical, Social and Governance Issues in AI. Congrats Vijay and Michael! Arxiv versions coming up soon!
10/2018: Posted slides of my Tutorial on Bias/Fairness in AI/ML geared towards the members of the European Parliament and privacy professionals! The talk video will be posted here (fpf.org/classes). Pictures from the event coming up soon!
09/2018: Our book chapter on “Computational Comparison of Metaheuristics” is online! Joint work with John Silberholz, Bruce Golden and Xingyin Wang.
09/2018: I am invited to speak in the panel session on Optimization in Power Systems, at the Power and Energy Systems Annual Meeting, August 2019.
09/2018: I am invited by the Future of Privacy Forum to give an introduction to Machine Learning and Fairness/Bias to members of the European parliament at their official side event at the 40th International Conference of Data Protection and Privacy Commissioners in Brussels (October 25th, 2018). Super excited to visit the European Parliament for the main event, as well as to speak at the side event! Stay tuned for the live stream of the event!
09/2018: Our paper on Robust Look-ahead Three-phase Balancing of Uncertain Distribution Loads got accepted in the Hawaii International Conference on System Sciences (HICSS-52)! Congratulations Le Xie and Xinbo!
09/2018: I will be giving an invited talk at the AMS Sectional Meeting (University of Hawaii at Manoa, March 22-24, 2019) and Mixed Integer Programming Workshop (MIP) 2019 (Sloan School of Management, MIT, July 2019).
09/2018: Sam Zhou qualified as a finalist for the INFORMS Undergraduate Operations Research Prize for our paper on Limited Memory Kelley’s Method (below), presentation @INFORMS Annual Meeting on Sunday, November 4th in the session SD26 (4:30pm in room 132A North Building)! Congratulations Sam!
09/2018: Our paper on Limited Memory Kelley’s Method Converges for Composite Convex and Submodular Objectives got accepted in NIPS 2018, with a spotlight presentation! Congratulations Madeleine and Sam!
07/2018: We arxived a new result with Madeleine Udell and Song Zhou: Limited Memory Kelley’s Method Converges for Composite Convex and Submodular Objectives! <convergence rate for novel ltd memory variant of Bach’s simpicial method –> solves Bach’s conjecture (2015), dual is also limited memory, variant of FCFW –> provably small subproblems, better run time!>
07/2018: Officially started as an Assistant Professor at Georgia Tech.
06/2018: I conducted an interactive session on Predictability and Learning during the Mission Possible Summer Camp for high school students at Georgia Tech! <combinatorial explosion is awesome, and yet we can learn>
04/2018: As the Microsoft Research Fellow, I will present a highlights of my work at the Industry Day, an all-day event at the Simons Institute with visitors from the companies that sponsor the institute.
04/2018: I am giving a talk in the next Simons Institute workshop on Mathematical and Computational Challenges in Real-time Decision Making on May 2, 2018.
04/2018: Our paper on “4/3 approximation for TSP on cubic 3-edge-connected graphs” has been accepted for publication in Operations Research Letters!
04/2018: I am visiting Microsoft Research, Redmond on April 23 and 24, 2018.
04/2018: I am giving a talk at the Stanford Theory Seminar on Thursday April 5, 4:15 PM, Gates 463A.
03/2018: I am giving a talk in the Session FB11: New Techniques in Discrete and Mixed Discrete Optimization, on February 23, Optimization Society Meeting, Denver.
03/2018: I am participating in the 1st Transatlantic-Transpacific Workshop of the ML Research Triangle, at Georgia Tech. I led the discussion on Grand Challenges in AI/ML, with my focus on Bias/Fairness in Optimization/ML.
02/2018: I am invited to the NSF workshop on Real-Time Learning and Decision Making in Dynamical Systems, Feb 12-13, Alexandria, VA.
02/2018: I will be presenting my research at the Discrete Optimization and Machine Learning Workshop in July, Tokyo.
02/2018: I am giving a talk on Learning Combinatorial Structures at Google Research, Mountain View.more
12/2017: Our book chapter on “Computational Comparison of Metaheuristics” is soon coming up! Joint work with John Silberholz, Bruce Golden and Xingyin Wang.
12/2017: “In Order Not to Discriminate, We Might Have to Discriminate“, an article by Christoph Drosser stemming from the discussions and talks at our Optimization and Fairness mini-symposium at the Simons Institute.
11/2017: I am giving a talk about “Learning What Works Best When” at Visa Research, Palo Alto.
11/2017: I am co-organizing an Optimization and Fairness Mini-symposium at the Simons Institute, along with David Williamson! Looking forward to a half-day of exciting talks!
11/2017: How can one explain their research to an average educated person? To meet this challenge, I will give a talk on “Alexa, What Should I Read Next?”, during the Fireside Chats competition at the Simons Institute.
10/2017: I am invited to give a talk at the Workshop on Algorithms and Optimization (January 2018), International Centre for Theoretical Sciences (ICTS), Bangalore.
09/2017: Our paper on “What Works Best When? A Systematic Evaluation of Heuristics for Max-Cut and QUBO” got accepted for publication in the INFORMS Journal on Computing!
08/2017: I will teach a module on the “Growth and Decay of Functions” at the Berkeley Math Circle (a weekly program for around 500 San Francisco area elementary, middle and high school students) in January, based off a BLOSSOMS video we shot earlier at MIT.
07/2017: I have been selected as the Microsoft Research Fellow, for the Real-Time Decision Making program at the Simons Institute!
05/2017: I have successfully defended my thesis titled, Combinatorial Structures in Online and Convex Optimization! I extend a heartfelt thanks to my advisors, committee members, friends and family for their encouragement and support.
01/2017: Our paper on “Newton’s Method for Parametric Submodular Function Minimization” got accepted for IPCO 2017!
12/2016: I am looking forward to attending a short winter course at the Harvard Law School on Internet & Society: The Technologies and Politics of Control, taught by Jonathan Zittrain and Joi Ito!
10/2016: What works best when? A Framework for Systematic Heuristic Evaluation (with John Silberholz, Iain Dunning) received a special recognition by INFORMS Computing Society in 2016 as a part of the student paper competition!
10/2016: An Efficient Algorithm for Dynamic Pricing using a Graphical Representation (with Maxime Cohen, Jeremy Kalas, Georgia Perakis) is a Finalist in the INFORMS Service Science Section student paper award 2016!