Events Calendar

PhD Dissertation Defense - Anand Balakrishnan

Title: From Qualitative to Quantitative Objectives for Neurosymbolic Control and Planning
 
Date and Time: Thursday, June 26, 2025 - 12:00p - 1:30p
 
Location: EEB 132 
 
Committee Members: Jyotirmoy Deshmukh (chair), Bhaskar Krishnamachari, Mukund Raghothaman, Erdem Biyik
 
 

Abstract: Reinforcement Learning (RL) is a popular paradigm by which an autonomous agent learns to perform complex tasks and behaviors through trial and error, facilitated by providing rewards to the agent. Effectively, these reward functions encode the high-level behavior intended by the designer, making the satisfactory performance of the tasks by the RL agent highly dependent on the reward functions. However, this raises concerns about safety and interpretability in the learned control policies. To this end, this dissertation proposes using formal specification paradigms that can express complex behaviors unambiguously, including time-dependent tasks like sequential tasks and patrolling tasks. This dissertation first presents how to extract quantitative rewards from such qualitative specifications without altering them. Through empirical and theoretical analysis, it also demonstrates the various guarantees and trade-offs associated with these techniques. Then, novel representations are derived for the specifications so that their structure can be directly exploited by optimization algorithms and leveraged to perform neurosymbolic control for complex systems.

 

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132

Audiences: Everyone Is Invited

Contact: Anand Balakrishnan

This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.

Add to Google Calendar