Truman Hickok
Independent researcher · robot learning & NeuroAI · Austin, TX
About
I'm an aspiring full-stack roboticist currently doing independent research that includes an accessible robot learning benchmark and an atlas for "the brain as an ML system". Previously, I was a research scientist at Southwest Research Institute and a graduate research assistant in the Neuromorphic AI Lab at UTSA.
Projects
A real2sim2real benchmark for evaluating robot foundation models on the SO-101 arm. 8,000 teleoperated demos, an Isaac Lab digital twin with sim2real transfer from sim data only, and VLAs, WAMs, and agents fine-tuned and evaluated under a matched compute budget. CoRL 2026 Late Breaking Results; poster at a CVPR 2026 workshop. Paper · Code · Data
A computational atlas of the brain as a learning system: an ML-style account of its architecture, learning rules ("optimizers" such as cortex and cerebellum), training data, and developmental curriculum.
Papers
-
SO-101 Bench: Accessible Spatial Manipulation
T. Hickok. CoRL 2026 Late Breaking Results. -
Scalable Strategies for Continual Learning with Replay
T. Hickok. arXiv 2025. -
Physics Informed Machine Learning (PIML) Methods for Estimating the Remaining Useful Lifetime (RUL) of Aircraft Engines
S. Nagaraj, T. Hickok. IEEE/AIAA Digital Avionics Systems Conference (DASC) 2024. -
BrowNNe: Brownian Nonlocal Neurons & Activation Functions
S. Nagaraj, T. Hickok. arXiv 2024. -
Watch Your Step: Optimal Retrieval for Continual Learning at Scale
T. Hickok, D. Kudithipudi. CVPR Workshops 2024. -
Optimal Retrieval for Continual Learning at Scale
T. Hickok. M.S. thesis, UT San Antonio, 2024.