Adaptive learning
Reinforcement learning, exploration, and generalization. How can agents acquire useful behavior and adapt when their observations or environments change?
Robust Policy Optimization →Research
Learning and reasoning for intelligent autonomy.
I develop reinforcement learning and foundation-model methods for intelligent systems that learn, reason, and act when conditions change.
My work connects adaptive decision making, multimodal reasoning, and human–robot collaboration. Robotics and medicine provide demanding settings for these questions.
Future research
Toward intelligent surgical autonomy.
I aim to connect adaptive learning, multimodal reasoning, and human–robot collaboration in bounded surgical tasks. This direction builds on my robotics and AI work, with clinical translation depending on appropriate partnerships and evaluation. The underlying questions extend to robot learning and intelligent decision systems beyond medicine.
Research
Three connected research themes
Reasoning for action
Language and multimodal representations for decision making. How can foundation models connect visual evidence and domain knowledge to useful actions?
Language-based representations →Collaborative autonomy
Robot skills, embodied interaction, and human–robot collaboration. How can learned skills support semi-autonomous systems with appropriate human intervention?
Surgical teleoperation →Selected contributions
Reinforcement learning · ARLET @ NeurIPS 2025 workshop
Robust Policy Optimization
My work on action-parameter perturbation studies policy optimization for reinforcement learning.
RPO is integrated into CleanRL and skrl. It is also used in Physical Intelligence’s released implementation of real-time action chunking.
Robust Policy Gradient Optimization through Action Parameter Perturbation in Reinforcement Learning
Language & learning · Findings of NAACL 2024
Language as a representation for learning
I investigate natural-language-based state representations in deep reinforcement learning, connecting semantic descriptions with sequential decision making.
Natural Language-based State Representation in Deep Reinforcement Learning
Multimodal reasoning · IAAI 2026, oral presentation
Reasoning from medical evidence
My work on multimodal burn assessment investigates how vision-language models can combine visual evidence and domain knowledge to support diagnostic reasoning.
Automated Unified Reasoning with Vision-Language Models for Multi-modal Burn Assessment
Human–robot collaboration · 2021
Semi-autonomous surgical teleoperation
My collaborative robotics work investigates surgical teleoperation and skill transfer, motivating systems that combine learned capabilities with human guidance.
SARTRES: A semi-autonomous robot teleoperation environment for surgery