Research and deployment
Selected work
Work on imitation and reward learning, curriculum learning, environment design, and robust learning under distribution shift.
Predictive Safety Curricula
Adaptive allocation of RL training experience using predictions of future safety cost. Evaluated in controlled simulation and physical-system deployment.
63% lower shank-collision incidence on ANYmal-D hardware across three matched training seeds.
Reward Learning under Distribution Shift
Uses variation across demonstrators as structural information to discourage reward features that fail under dynamics and nuisance shifts.
Evaluated reward transfer across five MuJoCo tasks under dynamics interventions.
Distribution-Matching Imitation Learning
Develops distribution-matching formulations of imitation learning, including sliced-Wasserstein objectives for occupancy matching.
Remained effective under up to 100× demonstration subsampling on Ant and Humanoid.
FASTRL: Reinforcement Learning in Surgical Digital Twins
Peer-reviewed work on reinforcement-learning benchmarks and assistance policies in surgical digital-twin environments.
Peer-reviewed benchmark and assistance-policy study in surgical digital twins.
Hyperbolic Wasserstein Autoencoders
Generative modeling with Wasserstein autoencoders in hyperbolic latent spaces.