Data-efficient, reliable vision-language-action (VLA) models for real-world humanoid robots — covering data collection, model development, and deployment. Physical AI is the wider thread: embodied systems that perceive, reason about, and act in the physical world with the reliability that real hardware demands.
Development of data-efficient, reliable VLA models for real-world humanoid robot use cases at Samsung SDS, spanning the full path from data collection through model development to deployment on physical platforms.
Pipelines for gathering and curating demonstration data for embodied learning, with an emphasis on getting strong policies out of modest data budgets rather than scaling collection indefinitely.
Grounding language instructions and visual observations in 3D scene understanding, drawing on multi-view 3D semantic segmentation and self-supervised representation learning for geometric data.
Quantifying policy uncertainty so an embodied system can tell a confident action from a guess — carrying the Bayesian and uncertainty-aware methods developed for vision and language models into control.
Closing the gap between benchmark performance and behavior on physical robots, where sensing noise, latency, and long-tail conditions determine whether a policy is actually usable.
Probabilistic and Gaussian Process modeling of physical processes and equipment, including uncertainty quantification and meta-modeling of asset digital twins — earlier work that informs how I approach physical-world models today.
The robotics work at Samsung SDS is largely unpublished. The publications below are the methodological foundations it builds on — uncertainty estimation, data-efficient 3D perception, and probabilistic modeling of physical systems.
Efficient uncertainty estimation for dense scene understanding — the perception side of embodied systems that must label and reason about their surroundings under a limited annotation budget.
Uncertainty-aware reward modeling for instruction following — directly relevant to VLA models, which must follow language instructions and know when their interpretation is uncertain.
Learning useful representations of 3D geometry without manual annotation, a building block for spatial understanding in embodied settings.
Gaussian Process modeling and optimization of a real physical deposition process, in collaboration with MIT — probabilistic modeling of physical systems with calibrated uncertainty.