Robert J. Moss
Scholar

Robert J. Moss

Google Scholar ID: OdCpu9sAAAAJ
Stanford University
decision making under uncertaintysafety validationPOMDP planningsurrogate modeling
Citations & Impact
All-time
Citations
583
 
H-index
10
 
i10-index
10
 
Publications
20
 
Co-authors
27
list available
Resume (English only)
Academic Achievements
  • Selected Publications: Kov: Transferable and Naturalistic Black-Box LLM Attacks using Markov Decision Processes and Tree Search (arXiv, 2024); ConstrainedZero: Chance-Constrained POMDP Planning Using Learned Probabilistic Failure Surrogates and Adaptive Safety Constraints (with Arec Jamgochian et al.); Contributed to 'Algorithms for Validation' (MIT Press, 2025).
Research Experience
  • Postdoc researcher at Stanford, part of SISL, Stanford Center for AI Safety, and Doerr School of Sustainability; Head TA for CS238/AA228 and CS238V/AA228V at Stanford; Previously a research staff member at MIT Lincoln Laboratory, core team member developing ACAS Xa, Xu, and sXu.
Education
  • PhD: Computer Science, Stanford University, 2025, advised by Mykel Kochenderfer; Master's: Computer Science, Stanford University, 2021, received the Christofer Stephenson Memorial Award.
Background
  • Research Interests: Safe planning, algorithms under uncertainty. Specialization: Computer Science, particularly in AI safety. Background: Focused on safe planning using surrogate models during his PhD at Stanford University.
Miscellany
  • Personal Interest: Loves Julia programming language.