Summary
Hyun Lee is a graduate researcher and computational neuroscientist with nine years of experience developing probabilistic and machine-learning models for neural time series and large-scale neural data. Based at Stanford after prior research at Columbia and Seoul National University, he builds novel switching autoregressive low-rank tensor models, spike-localization and registration pipelines, and image-based methods to track time-varying neural activity. His work blends principled statistical modeling with practical engineering—denoisers, point-neuron simulators, and multi-task neural nets—to improve inference accuracy and reproducibility across experiments. Comfortable spanning theory and systems, he has applied RL for model regularization, efficient CNN compression techniques, and LSTM attention for prognostics, revealing a knack for translating algorithmic ideas into scalable research tools. An ex-Sergeant and translator in the ROK Army, he brings discipline and cross-cultural communication skills to collaborative, interdisciplinary teams in Palo Alto.
9 years of coding experience