Xiangyu (Mike) Zhang

Senior Software Architecture Engineer
Ambarella Inc
Address 3101 Jay St
Santa Clara, CA 95054, United States
Contact xymzhang [AT] alumni.duke.edu

About me

Thanks for visiting my webpage! Currently, I am a software architecture engineer at Ambarella Inc, working on the performance of the next-generation AI chips. Explore our company website to find our state-of-art solutions for computer vision and autonomous driving! Meanwhile, I serve as a Technical Paper Committee member at SuperComputing 2022 (SC22) .

I received my Ph.D. in Electrical and Computer Engineering and M.S. in Computer Science at Duke University. My advisor is Prof. Alvin R. Lebeck. My research interests cover hardware accelerators for machine learning and statistical applications, statistical methods for hardware/system evaluation, and architecture with novel technologies. During my Ph.D. studies, I worked as an intern in Intelligence Driving Lab at Intel Labs China (Beijing, 2017) and at Ambarella Inc (Santa Clara, CA, 2018).

Before coming to Duke, I received my B.S. at South China Agricultural University. My undergraduate research focuses on non-linear optics, advised by Prof. Yongyao Li (currently in Foshan University, China) and Prof. Boris Malomed (currently in Tel Aviv University, Israel).

Education

2014 - 2020

Duke University

Doctor of Philosophy, Master of Science

Durham, NC, United States

2010 - 2014

South China Agricultural University

Bachelor of Science

Guangzhou, China

Publication

2021

Statistical Robustness of Markov Chain Monte Carlo Accelerators

Xiangyu Zhang, Ramin Bashizade, Yicheng Wang, Sayan Mukherjee, Alvin R. Lebeck

26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS ’21)
PDF  Slides 
2021

Optimizing Markov Random Field Inference via Event-Driven Gibbs Sampling on GPUs

Ramin Bashizade, Xiangyu Zhang, Sayan Mukherjee, Alvin R. Lebeck

CogArch21 Workshop
PDF 
2019

A Case for Quantifying Statistical Robustness of Specialized Probabilistic AI Accelerators

Xiangyu Zhang, Sayan Mukherjee, Alvin R. Lebeck

2019 IBM IEEE CAS/EDS – AI Compute Symposium, October 2019
arXiv 
2019

Rolling dice at the nanoscale

Xiangyu Zhang, Ramin Bashizade

ACM Crossroads, Fall 2019
Link to publisher  
2018

Architecting a stochastic computing unit with molecular optical devices

Xiangyu Zhang, Ramin Bashizade, Craig LaBoda, Chris Dwyer, Alvin R Lebeck

2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA)
Link to publisher   PDF   Talk slides   Lightening talk
2016

Accelerating Markov random field inference using molecular optical Gibbs sampling units

Siyang Wang, Xiangyu Zhang, Yuxuan Li, Ramin Bashizade, Song Yang, Chris Dwyer, Alvin R Lebeck

2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA)
[IEEE Micro's Top Picks 2016 Honorable Mention]
Link to publisher   PDF   Talk slides
2015

Discrete solitons in self-defocusing systems with PT-symmetric defects

Zhiqiang Chen, Jiasheng Huang, Jinglei Chai, Xiangyu Zhang, Yongyao Li, Boris A Malomed

Physical Review A
Link to publisher
2015

All-optical diode realized by one-dimensional waveguide array

Zhenjing Peng, Hongji Li, Zhiwei Fan, Xiangyu Zhang, Yongyao Li

Journal of Nonlinear Optical Physics & Materials
Link to publisher
2015

Transmission, reflection, scattering, and trapping of traveling discrete solitons by C and V point defects

Jin-Hong Huang, Hong-Ji Li, Xiang-Yu Zhang, Yong-Yao Li

Frontiers of Physics
Link to publisher
2014

Discrete solitons and scattering of lattice waves in guiding arrays with a nonlinear PT-symmetric defect

Xiangyu Zhang, Jinglei Chai, Jiasheng Huang, Zhiqiang Chen, Yongyao Li, Boris A Malomed

Optics Express
Link to publisher   arXiv
2014

Antisymmetry breaking of discrete dipole gap solitons induced by a phase-slip defect

Xiangyu Zhang, Jinglei Chai, Dezhao Ou, Yongyao Li

Modern Physics Letters B
Link to publisher