Chester Huynh

Chester Huynh

Software Engineer @ Microsoft

Publications

Analysis of Neural Fragility: Bounding the Norm of a Rank-One Perturbation Matrix

Published in arXiv Machine Learning, 2022

This paper studies the analytical properties of the rank-one perturbation matrices in linear dynamical systems. This provides additional analysis on neural fragility, which is a recently proposed model by Li et al. (Nature Neuroscience, 2021) predicting surgical outcome for epilepsy patients.

Recommended citation: Li, A., Huynh, C.. Analysis of Neural Fragility: Bounding the Norm of a Rank-One Perturbation Matrix (2022). https://doi.org/10.48550/arXiv.2202.07026

Manifold Oblique Random Forests: Towards Closing the Gap on Convolutional Deep Networks

Published in arXiv Machine Learning, 2021

This paper proposes a manifold-aware variation of the random forest algorithm for structured data, which has empirically demonstrated high statistical efficiency relative to deep neural networks.

Recommended citation: Perry, R., Li, A., Huynh, C. et al. Manifold Oblique Random Forests: Towards Closing the Gap on Convolutional Deep Networks. arXiv Machine Learning (2019). https://doi.org/10.48550/arXiv.1909.11799

Towards Automatic Localization and Anatomical Labeling of Intracranial Depth Electrodes in Brain Images

Published in Engineering in Medicine and Biology Conference, 2020

This paper outlines the pipeline for semi-automatically localizing and labeling SEEG depth electrodes in 3D images of brain volumes. Fully automating this task is left to future work.

Recommended citation: Huynh C, Li, A Gonzalez-Martinez J, Sarma SV. Towards Automatic Localization and Anatomical Labeling of Intracranial Depth Electrodes in Brain Images. Conf Proc IEEE Eng Med Biol Soc. 2020. https://sarmalab.icm.jhu.edu/publications/