Yan Zhang, Ph.D.
Lawrence Berkeley National Laboratory
1 Cyclotron Road
Bldg 59, Room 4024EBerkeley, CA 94720 US
Ph.D., Electrical Engineering, Northeastern University, Boston, MA
B.E., Automation/Control Engineering, Beijing University of Technology, China
ExperienceNERSC Postdoctoral Fellow, Lawrence Berkeley National Laboratory, Berkeley, CA
- Worked on Deep Learning at scale (parallel/distributed), extreme-scale spatial-temporal learn- ing algorithms for multimodal data streams (fMRI brain image) using modal/data parallelism.
- Analyzed kernel performance of deep neural network applications based on Roofline model and consulted HPC users for optimization strategies/approaches using NVIDIA profiling tools.
- Served as reviewer for Deep Learning & Data Analytics related scientific research proposals of using the next generation heterogeneous CPUs-GPUs supercomputing system - Perlmutter.
- Postdoctoral Appointee, Argonne National Laboratory, Lemont, IL
- Designed deep/transfer learning models (Convolutional Neural Networks) for classification, regression, anomaly detection and feature extraction on electron microscopy/spectroscopy.
- Developed smart imaging and dynamic sampling methods using both supervised and unsupervised learning to reduce acquisition time and radiation for scanning imaging modalities.
- Applied convolutional auto-encoders, generative adversarial networks (GANs) and neural style transfer for electron microscopic image synthesis using parallel and distributed GPUs processing.
Graduate Research Assistant, Northeastern University, Boston, MA
- Designed classification (SVM, Neural Networks) and regression (Least-square, Lasso, Ridge, Bayesian) models for x-ray scattering data to recognize shape and size of macromolecules.
- Developed clustering and image processing algorithms to analyze XRD images of biological tissue - feature selection, dimensionality reduction, pattern matching and spectral clustering.
- Applied data quantization, pair-distance function and Fourier method to simulate diffraction patterns of macromolecules and to build dictionary for parameter estimation of measurements.
- Developed image analysis and pattern recognition algorithms for components detection as well as contaminants and artifacts on AFM images of hard drive recording heads.
- Developed computer vision and machine learning algorithms for automatic defect classification on SEM images of semiconductor wafer - edge detection, segmentation in C++.
- Father of Molecular-MNIST (2019), arXiv:1911.07644
- ATPESC (2019), Argonne Training Program on Extreme-Scale Computing, St. Charles, IL
- Dissertation Completion Fellowship Award (2016), Northeastern University, Boston, MA
- First-Class Scholarship for Academic Excellence (2009), Beijing University of Technology, China
- Y. Zhang, et al. (2018) “Reduced electron exposure for energy-dispersive spectroscopy using dynamic sampling”. Ultramicroscopy, 184:90-97.
- Y. Zhang, et al. (2016) “Diffraction pattern simulation of cellulose fibrils using distributed and quantized pair-distances”. Journal of Applied Crystallography, 49(6):2244-2248.
- Y. Zhang, et al. (2015) “Breakdown of hierarchical architecture in cellulose during dilute acid pretreatments”. Cellulose, 22(3):1495-1504.
- H. Inouye, Y. Zhang, et al. (2014) “Multiscale deconstruction of molecular architecture in corn stover”. Scientific Reports, 4, 3756
- Y. Zhang, et al. (2019) “A Molecular-MNIST Dataset for Machine Learning Study on Diffraction Imaging and Microscopy”. arXiv:1911.07644.
- Y. Zhang, et al. (2018) “U-SLADS: UnSupervised Learning Approach for Dynamic Dendrite Sampling”. arXiv:1807.02233.
- Y. Zhang, et al. (2018) “ SLADS-Net: Supervised Learning Approach for Dynamic Sampling using Deep Neural Networks”. IS&T International Symposium on Electronic Imaging.
- Y. Zhang, et al. (2017) “Deep learning, dynamic sampling and smart energy-dispersive spec- troscopy”. Frontiers in Optics, OSA.
- Y. Zhang, et al. (2017) “Under-sampling and image reconstruction for scanning electron micro- scopes”. Microscopy and Microanalysis, MSA.
- Y. Zhang, et al. (2015) “A new pre-processing method for scanning x-ray microdiffraction patterns”. IEEE Biomedical Circuits and Systems Conference.
- Y. Zhang and L. Makowski (2015) “Auto-thresholding edge detector for bio-image processing”. The 41st Northeast Bioengineering Conference, IEEE.
- L. Yu, Y. Zhang, et al. (2015) “High performance computing of fiber scattering simulation”. The 8th Workshop on General Purpose Processing using GPUs, ACM.
- Y. Zhang, et al. (2014) “Fast simulation of x-ray diffraction patterns from cellulose fibrils using GPUs”. The 40th Northeast Bioengineering Conference, IEEE.