Pengmiao Zhang
Computer Engineering, USC
pengmiao@usc.edu
Hello,
I am Pengmiao Zhang (张鹏淼), a PhD candidate in Ming Hsieh Department of Electrical and Computer Engineering at University of Southern California.
I am studying and working as a Research Assistant in the Data Science Lab under the supervision of Professor Viktor K. Prasanna.
I focus on the research of efficient machine learning for memory access prediction and data prefetching, targeting practical AI-empowered computer architectures.
Education
PhD in Computer Engineering, 2019 - Present
University of Southern California, Los Angeles, USA
M.Eng in Electrical Engineering, 2013 - 2015
Harbin Institute of Technology, Harbin, China
B.Sc in Electrical Engineering, 2009 - 2013
Northeastern University, Shenyang, China
Work Experience
University of Southern California
Research Assistant, 2019 - Present
Teaching Assistant: EE-450 Introduction to Computer Networks , Fall 2020 - Spring 2024
Teaching Assistant: EE-557 Computer Systems Architecture , Fall 2021
Teaching Assistant: EE-355 Software Design for Electrical Engineers, Spring 2020
Neusoft Corporation, Shenyang, China
Research Scientist, 2018 - 2019
Shenyang Engine Design Institute, Shenyang, China
Electrical Engineer, 2015 - 2018
Publications
Gupta, Neelesh, Narayanan Kannan, Pengmiao Zhang, Rajgopal Kannan, and Viktor K. Prasanna. “TabConv: Low-Computation CNN Inference via Table Lookups”. (Under Review)
[IPDPS '24] Zhang, Pengmiao, Neelesh Gupta, Rajgopal Kannan, and Viktor K. Prasanna. “Attention, Distillation, and Tabularization: Towards Practical Neural Network-Based Prefetching”. IEEE International Parallel & Distributed Processing Symposium (IPDPS), 2024. (Accepted)
[SC '23] Zhang, Pengmiao, Rajgopal Kannan, and Viktor K. Prasanna. “Phases, Modalities, Temporal and Spatial Locality: Domain Specific ML Prefetcher for Accelerating Graph Analytics,” The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC ’23), 2023. (Acceptance rate: 23.9%)
[HPEC '23] Gupta, Neelesh, Pengmiao Zhang, Rajgopal Kannan, and Viktor K. Prasanna, "PaCKD: Pattern-Clustered Knowledge Distillation for Compressing Memory Access Prediction Models", 2022 IEEE High Performance Extreme Computing Virtual Conference (HPEC ’23), 2023.
[HPEC '23] Gorle, Abhiram Rao, Pengmiao Zhang, Rajgopal Kannan, and Viktor K. Prasanna, "G-MAP: A Graph Neural Network-Based Framework for Memory Access Prediction", 2022 IEEE High Performance Extreme Computing Virtual Conference (HPEC), 2023.
[HiPC '23] Marino, Kyle, Pengmiao Zhang, and Viktor K. Prasanna, "ME-ViT: A Single-Load Memory-Efficient FPGA Accelerator for Vision Transformers", 30th IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC), 2023. (Acceptance rate: 24%) (Best Paper Award)
[SC'22] Zhang, Pengmiao, Rajgopal Kannan, Ajitesh Srivastava, Anant V. Nori, and Viktor K. Prasanna. "ReSemble: reinforced ensemble framework for data prefetching." In 2022 SC22: International Conference for High Performance Computing, Networking, Storage and Analysis (SC), pp. 1168-1181. IEEE Computer Society, 2022.
[HPEC '22] Zhang, Pengmiao, Rajgopal Kannan, Xiangzhi Tong, Anant V. Nori, and Viktor K. Prasanna. "SHARP: Software Hint-Assisted Memory Access Prediction for Graph Analytics." In 2022 IEEE High Performance Extreme Computing Conference (HPEC), pp. 1-8. IEEE, 2022.
[DATA '22] Zhang, Pengmiao, Rajgopal Kannan, Anant Nori, and Viktor Prasanna. "A2P: Attention-based Memory Access Prediction for Graph Analytics." In the 11th International Conference on Data Science, Technology and Applications. 2022.
[CF '22] Zhang, Pengmiao, Ajitesh Srivastava, Anant V. Nori, Rajgopal Kannan, and Viktor K. Prasanna. "Fine-grained address segmentation for attention-based variable-degree prefetching." In Proceedings of the 19th ACM International Conference on Computing Frontiers, pp. 103-112. 2022.
[ISCA '21] Zhang, Pengmiao, Ajitesh Srivastava, Kannan, Rajgopal Kannan, Anant V. Nori, Viktor K. Prasanna, "TransforMAP: Transformer for Memory Access Prediction." The International Symposium on Computer Architecture (ISCA), ML for Computer Architecture and Systems Workshop, 2021
[JDSA] Zhang, Pengmiao, Ajitesh Srivastava, Ta-Yang Wang, Cesar AF De Rose, Rajgopal Kannan, and Viktor K. Prasanna. "C-MemMAP: Clustering-driven Compact, Adaptable, and Generalizable Meta-LSTM Models for Memory Access Predictio." International Journal of Data Science and Analytics (2021): 1-14.
[MemSys '20] Zhang, Pengmiao, Ajitesh Srivastava, Benjamin Brooks, Rajgopal Kannan, and Viktor K. Prasanna. "RAOP: Recurrent Neural Network Augmented Offset Prefetcher." In The International Symposium on Memory Systems, pp. 352-362. 2020
[PAKDD '20] Srivastava, Ajitesh, Ta-Yang Wang, Pengmiao Zhang, Cesar Augusto F. De Rose, Rajgopal Kannan, and Viktor K. Prasanna. "MemMAP: Compact and Generalizable Meta-LSTM Models for Memory Access Prediction ." Advances in Knowledge Discovery and Data Mining 12085 (2020): 57.
Jin, Miaoxin, Pengmiao Zhang, Gang Li, Qiang Gao, Xiaolu Li, and Dianguo Xu. "A downhole multi-parameter monitoring system for electrical submersible pump." In 2015 9th International Conference on Power Electronics and ECCE Asia (ICPE-ECCE Asia), pp. 2427-2431. IEEE, 2015.
Zhang, Pengmiao, Qiang Gao, Liang Cheng, and Dianguo Xu. "Research on downhole multi-parameter comprehensive measurement of ESP." In 2014 Fourth International Conference on Instrumentation and Measurement, Computer, Communication and Control, pp. 346-350. IEEE, 2014.
Contact
Address: 3740 McClintock Ave, EEB 236, Los Angeles, CA 90089
E-Mail: pengmiao AT usc.edu