Zongwei Zhou

Zongwei Zhou

Carnegie Mellon University

H-index: 19

North America-United States

About Zongwei Zhou

Zongwei Zhou, With an exceptional h-index of 19 and a recent h-index of 17 (since 2020), a distinguished researcher at Carnegie Mellon University, specializes in the field of Machine learning performance, hardware accelerator co-design, compiler, computer system and security.

His recent articles reflect a diverse array of research interests and contributions to the field:

Gemini: a family of highly capable multimodal models

Tpu v4: An optically reconfigurable supercomputer for machine learning with hardware support for embeddings

Palm: Scaling language modeling with pathways

Overlap communication with dependent computation via decomposition in large deep learning models

Glam: Efficient scaling of language models with mixture-of-experts

Bigssl: Exploring the frontier of large-scale semi-supervised learning for automatic speech recognition

AutoDistill: An end-to-end framework to explore and distill hardware-efficient language models

Sparsely Activated Language Models are Efficient In-Context Learners

Zongwei Zhou Information

University

Position

Ph.D. Candidate of Electrical and Computer Engineering

Citations(all)

5270

Citations(since 2020)

4032

Cited By

1502

hIndex(all)

19

hIndex(since 2020)

17

i10Index(all)

24

i10Index(since 2020)

21

Email

University Profile Page

Google Scholar

Zongwei Zhou Skills & Research Interests

Machine learning performance

hardware accelerator co-design

compiler

computer system and security

Top articles of Zongwei Zhou

Tpu v4: An optically reconfigurable supercomputer for machine learning with hardware support for embeddings

2023/6/17

Overlap communication with dependent computation via decomposition in large deep learning models

2022/12/19

Glam: Efficient scaling of language models with mixture-of-experts

2022/6/28

Bigssl: Exploring the frontier of large-scale semi-supervised learning for automatic speech recognition

IEEE Journal of Selected Topics in Signal Processing

2022/6/13

AutoDistill: An end-to-end framework to explore and distill hardware-efficient language models

arXiv preprint arXiv:2201.08539

2022/1/21

Sparsely Activated Language Models are Efficient In-Context Learners

2022

Method and apparatus for trusted display on untrusted computing platforms to secure applications

2021/12/14

Ten lessons from three generations shaped google’s tpuv4i: Industrial product

2021/6/14

See List of Professors in Zongwei Zhou University(Carnegie Mellon University)

Co-Authors

academic-engine