I am a final-year PhD student in Computer Science at the University of Southern California School of Engineering (USC-Viterbi) , under the supervision of Professor Jyotirmoy Deshmukh.

My research area lies at the intersection of Artificial Intelligence and Temporal Logics, with applications in Formal Verification of Learning Enabled Systems and Neurosymbolic Reinforcement Learning.

Selected Experience

Open Source Contribution

I generated the first deterministic formal verification framework for Signal Temporal logics in collaboration with research scientists of Toyota Research Institute North of America (TRINA) STLVerNN that verifies general Temporal specifications for Learning enabled autonomous systems.

I scaled the process of Neural Feedback Policy Learning for agents to satisfy Signal Temporal Logics specification STL_dropout. This was achieved by combination of stochastic depth (firstly proposed for ResNet) via Neurosymbolic MBRL.

Research Outcomes

I have published 4 jouranal papers and 15 conference papers focusing on Neurosymbolic Learning and Verification, CPS and Control Theory in reputable IEEE and ACM journals and conferences. Feel free to see my publication entry

Upcomming News

1- My paper : LB4TL: Smooth Semantics for Temporal Logic for Scalable Training of Neural Feedback Controllers.
has been published as an IFAC conference paper.
2- My paper : Scaling Learning based Policy Optimization for Temporal Tasks via Dropout
has been accepted in ACM Tranactions of Cyber Physical Systems (TCPS 2024).
3- My paper : Statistical Reachability Analysis of Stochastic Cyber-Physical Systems under Distribution Shift
has been accepted in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD 2024), and will be presented in ESWeek as a part of EMSOFT conference.