I'm a computer science PhD student at Stanford University studying artificial intelligence. I completed a master's in CS from Stanford and received the Christofer Stephenson Memorial Award for the best CS master's thesis. My current research is on black-box validation of safety-critical autonomous systems using reinforcement learning, deep learning, and Bayesian optimization and I'm interested in large-scale POMDP planning for decision making under uncertainty using surrogate models. Oh, and I love Julia.
I'm currently a grad student researcher at the Stanford Intelligent Systems Laboratory (SISL) advised by Mykel Kochenderfer and a part of the Stanford Center for AI Safety and the Stanford Doerr School of Sustainability (SCERF and Mineral-X). I was the head TA for Stanford's CS238/AA228: Decision Making Under Uncertainty and a course development assistant for AA120Q: Building Trust in Autonomous Systems and subsequently won the Centennial TA Award recognizing my teaching efforts.
Before Stanford, I was a research staff member at MIT Lincoln Laboratory and was part of the core team that developed, optimized, and validated the next-generation aircraft collision avoidance system, certified by the FAA (ACAS Xa, Xu, and sXu).

Thesis

Algorithms for efficient validation of black-box systems

Robert J. Moss

Master's thesis, Stanford University, 2021

Teaching

Decision Making Under Uncertainty using POMDPs.jl

Julia Academy

Publications (Selected)

BetaZero: Belief-State Planning for Long-Horizon POMDPs using Learned Approximations

Robert J. Moss, Anthony Corso, Jef Caers, and Mykel J. Kochenderfer

Under review (2023)

Bayesian Safety Validation for Black-Box Systems

Robert J. Moss, Mykel J. Kochenderfer, Maxime Gariel, and Arthur Dubois

AIAA AVIATION Forum (2023)

A Survey of Algorithms for Black-Box Safety Validation of Cyber-Physical Systems

Anthony Corso, Robert J. Moss, Mark Koren, Ritchie Lee, Mykel J. Kochenderfer

Journal of Artificial Intelligence Research (JAIR), 2021

Autonomous Vehicle Risk Assessment

Robert J. Moss, Shubh Gupta, Robert Dyro, Karen Leung, Mykel J. Kochenderfer, Grace X. Gao, Marco Pavone, Edward Schmerling, Anthony Corso, Regina Madigan, Matei Stroila, and Tim Gibson

Stanford Center for AI Safety, 2021

Predictive Risk for Efficient Black-Box Validation of Autonomous Vehicles

Robert J. Moss

Stanford University (CS229: Machine Learning), 2021

Cross-Entropy Method Variants for Optimization

Robert J. Moss

arXiv, 2020

Adaptive Stress Testing of Trajectory Predictions in Flight Management Systems

Robert J. Moss, Ritchie Lee, Nicholas Visser, Joachim Hochwarth, James G. Lopez, and Mykel J. Kochenderfer

Digital Avionics Systems Conference (2020)

ACAS Xu: Integrated Collision Avoidance and Detect and Avoid Capability for UAS

Michael P. Owen, Adam Panken, Robert Moss, Luis Alvarez, and Charles Leeper

Digital Avionics Systems Conference (2019)

Bayesian Network Model of Pilot Response to Collision Avoidance Resolution Advisories

Edward H. Londner and Robert J. Moss

Journal of Air Transportation, Volume 26, Number 4 (2018), pp. 171–182

Automated Dynamic Resource Allocation for Wildfire Suppression

J. Daniel Griffith, Mykel J. Kochenderfer, Robert J. Moss, Velibor V. Mišić, Vishal Gupta, and Dimitris Bertsimas

Lincoln Laboratory Journal, Volume 22, Issue 2, pp. 38–59, 2017

Rotation Curve for the Milky Way Galaxy in Conformal Gravity

James G. O'Brien and Robert J. Moss

Journal of Physics: Conference Series, Volume 615 012002, 2015

Using Julia as a Specification Language for the Next-Generation Airborne Collision Avoidance System

Robert J. Moss

JuliaCon, 2015