-
PhD Thesis Proposal - Bryon Tjanaka
Wed, Apr 16, 2025 @ 03:30 AM - 05:00 PM
Thomas Lord Department of Computer Science
University Calendar
Title: Robust Robots by Scaling Up Quality Diversity Optimization
Date: April 16 at 3:30-5:00PM
Location: GCS 402C
Committee:
Chair: Stefanos Nikolaidis
Erdem Biyik
Gaurav Sukhatme
Willie Neiswanger
External (from ECE): C.-C. Jay Kuo
Abstract: Quality diversity (QD) optimization is a paradigm for finding diverse, high-performing collections of solutions. QD offers a promising approach for developing robots that can adapt to a wide variety of circumstances. In this proposal, I will discuss my work on scaling up QD to handle the demands of larger robotics problems. First, I will discuss several works that scale up QD to optimize neural network policies in reinforcement learning. These works form a toolbox for practitioners looking to apply QD to their work in robotics. Next, I will discuss my proposed work on scaling up QD to account for more flexible user preferences. I conclude by discussing several other aspects of my research including environment generation and the Pyribs library of QD algorithms
Location: Ginsburg Hall (GCS) - 402C
Audiences: Everyone Is Invited
Contact: Bryon Tjanaka
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.
