Suguman Bansal

I am a CRA/NSF Computing Innovation Postdoctoral Fellow with Prof. Rajeev Alur at the Department of Computer and Information Sciences at University of Pennsylvania. I completed my Ph.D under Prof. Moshe Y. Vardi at Rice University.

I am on the Academic Faculty Job Market.
Application Materials: [Research Statement] [Teaching Statement] [Diversity Statement] [Resume]

My research vision is to make AI trustworthy by close integration of formal methods and programming languages with artificial intelligence and machine learning. Towards trustworthy computing, my research has focused on simplifying the design-development-deployment pipeline of safe and reliable AI systems. I have worked on automated controller synthesis, i.e., the automated generation of control systems from a given high-level specification with theoretical guarantees of correctness. My current research thrusts include:

Formal Methods for Reinforcement Learning [CAV22, NeurIPS21, Invited]

Automata-Based Quantitative Reasoning [AAAI22, TACAS21, CAV19, CAV18b, FoSSaCS18]

Reactive Synthesis from High-Level Specifications [CAV21, POPL20, AAAI20, CAV18a]

Awards and Honors

MIT EECS Rising Star, 2021

CRA/NSF Computing Innovation Fellow, 2020

Future Faculty Fellow 2019, Rice School of Engineering, 2019

MIT EECS Rising Star, 2018

Invited to Simons Institute for program on Real-time decision making, Spring 2018

Invited to the 1st Google Student Research Summit 2017, September 2017

Invited to Dagsthul Seminar on Game Theory, AI, Logic and Algorithms, March 2017

One of the 10 graduate students invited to Microsoft Research Faculty Summit 2016, July 2016

Gold Medal, ACM Student Research Competition at POPL 2016, 2016

Andrew Ladd Graduate Fellowship, 2015 - 2016

CMI Undergraduate Scholarship, 2011 - 2014

News

June 6, 2022I am the Round-Table Host on "Safe Reinforcement Learning" at WOLVERINE at FLoC 2022. Do attend if you want to learn about the cutting edge reserach at the intersection of Formal Methods and Reinforcement Learning.

April 30, 2022New CAV Paper! Specification-Guided Learning of Nash Equilibria with High Social Welfare has been accepted to CAV 2022. This is joint work with Kishor Jothimurugan, Osbert Bastani, and Rajeev Alur.

December 9, 2021Paper on Compositional Reinforcement Learning from Logical Specifications appears at NeurIPS 2021 today. This is joint work with Kishor Jothimurugan, Osbert Bastani, and Rajeev Alur.

December 1, 2021Paper on On Synthesis from Satisficing and Temporal Goals has been accepted to appear at AAAI 2022. This is joint work with Lydia Kavraki, Moshe Y. Vardi, and Andrew Wells. [10-mins Talk Video]

October 19, 2021Paper on Specification-Guided Learning of Nash Equilibria with High Social Welfare has been accepted to appear at SafeRL 2021 at NeurIPS 2021

All news…