Caleb Miller is an applied mathematician and data scientist in the Center for Applied Scientific Computing in the Data Science and Analytics Group. His research interests include Bayesian statistics, Monte Carlo methods, importance sampling, reinforcement learning, optimal control, data fusion, optimization, and game theory.

Caleb first interned at LLNL in 2019 working on importance sampling methods for rare events in the space domain. After receiving his Ph.D. from the University of Colorado, Boulder in applied mathematics he joined the lab as a machine learning postdoctoral researcher focusing on reinforcement learning for decision making. In 2025 he became full-time staff at LLNL with projects that support Strategic Deterrence and arms control.

PH.D Applied Mathematics, University of Colorado, Boulder

M.S/B.S Mathematics, California Polytechnic State University, San Luis Obispo

Selected:

Soper, B. C., Miller, C. J., Merl, D. M. (2024, September). Linearly Solvable General-Sum Markov Games. In 2024 60th Annual Allerton Conference on Communication, Control, and Computing (pp. 1-8). IEEE.

Miller, Caleb J., Braden C. Soper, Amanda Muyskens, Benjamin W. Priest, Michael D. Schneider, and Dan M. Merl. Exploration with Scalable Gaussian Process Reinforcement Learning. No. LLNL-TR-860762. Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States), 2024.

Miller, C., Corcoran, J. N., & Schneider, M. D. (2021). Rare events via cross-entropy population monte carlo. IEEE Signal Processing Letters, 29, 439-443.