BansalLab

[Publications] [Tutorial] [Group Members] [Open Positions]

BansalLab is located at the School of Computer Science at Georgia Institute of Technology.
We work on building Trustworthy AI by simplifying the design-development-deployment pipeline of safe and reliable AI systems. We apply Formal Methods and Logic to Artificial Intelligence, Machine Learning, and Programming Languages. Our current research thrusts include:

  • Formal Methods for Reinforcement Learning [SAS22Keynote, AAAI23Tutorial, CAV22, NeurIPS21, Henzinger60]
  • Symbolic Quantitative Reasoning [IJCAI23, AAAI22, TACAS21, CAV19, CAV18b, FoSSaCS18]
  • Reactive Synthesis from High-Level Specifications [ATVA23 (Best Paper), CAV21, POPL20, AAAI20, CAV18a]
  • Group Members

    Faculty
  • Suguman Bansal

  • PhD Students
  • Vignesh Subramanian

  • Masters and Undergraduate Students
  • Kaushik Arcot (MS)
  • Sahaj Purohit (BS)
  • Ramneet Singh (IIT Delhi) (MS/BS)
  • Siddharth Meenachi Sundaram (MS)

  • Alumni
  • Yash Kankariya (BS)
  • Open Positions

    We are growing. We are looking for highly motivated graduate students (PhD and Masters), postdocs, and exceptional undergraduate students. We work at the intersection of theory and algorithmic design; therefore, an ideal candidate should have deeper expertise in one area and willingness to learn the other. A strong background in algorithms/formal methods, statistics, and prior experience in coding is crucial to make a significant contribution to our research.

    We are growing. Please get in touch and send me your CV, if interested in these topics:
    (a) Theory and practice of Formal Methods in Reinforcement Learning (RL),
    (b) Automata and Logical Methods.
    Additional Instructions for Prospective Group Members:
    1. Phd and Masters: Apply To GaTech by December 15.
    2. Georgia Tech Masters and Undergraduates: Must be willing to commit at least one full semester of work. Preferably, you should have taken at least one course with me. Otherwise, you should read and summarize at least one of my paper.
    3. Posdocs: One postdoc position at the intersection of logic and RL. Let us chat.

    Recent Group News