Dual Objective Reinforcement Learning with Novel HJB Forms - Recently Submitted, September 2025
MADR: MPC-guided Adversarial DeepReach - Recently Submitted, September 2025
Reachability Barrier Networks: Learning HJ for Smooth and Flexible CBFs - Recently Submitted, April 2025
Linear Supervision for Nonlinear, High-Dimensional Neural Control and Differential Games - Learning for Dynamics & Control (Best Paper Nom.) - December 2024
State-Augmented Linear Games with Antagonistic Error for High-Dimensional, Nonlinear Hamilton-Jacobi Reachability - Conference on Decision and Control - March 2024
Conservative Linear Envelopes for High-Dimensional, Hamilton-Jacobi Reachability for Nonlinear Systems via the Hopf Formula - Transactions on Automatic Control - April 2024
Research Experience
Visiting Chuchu Fan's REALM lab (Fall 2025).
Education
Undergraduate degree in Applied Mathematics from UC Berkeley, where he worked with Claire Tomlin, Adam Arkin, and Jay Keasling; currently pursuing a PhD at UC San Diego under Sylvia Herbert's Safety and Autonomous Systems group.
Background
Mechanical and Aerospace PhD candidate, focusing on differential games (HJB-PDE), control, and learning. Thesis concentrates on feasible methods for safe, high-dimensional autonomy. HHMI/NIH Interfaces fellow, supported by ONR and the Society of Hellman.
Miscellany
Contact: willsharpless [at] ucsd [dot] edu / CV / Google Scholar / Github / Linkedin