|
Xiaowen Chu
(PhD, HKUST, 2003) Professor
(No-Pay Leave Now) Kowloon Tong, Kowloon, HONG KONG |
Brief Biography |
Dr.
Chu received his B.Eng. degree in the Computer Science from Tsinghua University, Beijing, P. R.
China, in 1999, and the Ph.D. degree in the Computer Science from the Hong Kong University of Science and Technology
in 2003. He is a professor in the Department
of Computer Science, Hong Kong Baptist
University. He is serving as the Director of the High Performance Cluster Computing
Centre of HKBU, the Director of Blockchain and Fintech Laboratory,
and the Director of High
Performance Machine Learning Laboratory. He is a senior member of IEEE
and a member of ACM. He is a vice-chairman of the Blockchain Technical
Committee of China Institute
of Communications. |
|
|
Job
Opening |
Post-doctoral
Research Fellow in High Performance Deep Learning A
Postdoctoral position is immediately available in the Department of Computer
Science at Hong Kong Baptist University, Hong Kong. Applicants should possess
a PhD Degree in computer science, computer engineering, or a related field,
and sufficiently demonstrate abilities to conduct high-quality research in
the area of high performance deep learning. Initial duration of this position
is 12 months, and is renewable subject to satisfactory performance and mutual
agreement. We offer a competitive salary and benefits package. Hong Kong
practises a simple
and low-rate tax system. Interested applicants are invited to send
a CV to Dr. Xiaowen Chu by email (chxw@comp.hkbu.edu.hk). The position remains
open until it is filled. Research
Assistant / Senior Research Assistant in High Performance Deep Learning Research
assistant positions are immediately available in the Department of Computer
Science at Hong Kong Baptist University, Hong Kong. Applicants should possess
a Bachelor Degree (or equivalent) in computer science, computer engineering,
or a related field. R&D experience in any of the following areas is
considered as a plus: (i) GPU computing, (ii) Deep learning, (iii) Network
performance modelling. Initial duration of this position is 12 months, and is
renewable subject to satisfactory performance and mutual agreement. We offer a
competitive salary and benefits package.
Interested applicants are invited to send a CV to Dr. Xiaowen Chu by
email (chxw@comp.hkbu.edu.hk).
The position remains open until it is filled. |
News |
[14 May
2021] Our
paper “Exploiting Simultaneous Communications to Accelerate Data Parallel
Distributed Deep Learning” has received the Best Paper Award of IEEE INFOCOM 2021. Congratulations to
Dr. SHI Shaohuai and Prof. Bo Li! A preprint can be found at arXiv. [11 April
2021] The
paper “BU-Trace: A Permissionless Mobile System for Privacy-Preserving
Intelligent Contact Tracing” has received the Best Paper Award of the 2021
International Workshop on Mobile Ubiquitous Systems and Technologies,
collated with DASFAA 2021. Congratulations
to all team members! A preprint can be found at arXiv. [29 March
2021] The
paper “P2B-Trace:
Privacy-Preserving Blockchain-based Contact Tracing to Combat Pandemics” has
been accepted by SIGMOD 2021.
Congratulations to all team members! A preprint can be found at arXiv. [6 March
2021] The
paper “IRS: A Large Naturalistic
Indoor Robotics Stereo Dataset to Train Deep Models for Disparity and Surface
Normal Estimation” has been accepted by ICME
2021. Congratulations to all team members! A preprint can be found at arXiv. [4 March
2021] The
paper “EDNet: Efficient
Disparity Estimation with Cost Volume Combination and Attention-based Spatial
Residual” has been accepted by CVPR
2021. This work is collaborated with Tongji University. Congratulations
to all team members! A preprint can be found at arXiv. [14 Jan
2021] The
paper “MG-WFBP: Merging Gradients Wisely for Efficient Communication in
Distributed Deep Learning” has been accepted by IEEE TPDS. Congratulations to Dr. SHI Shaohuai! [26 Dec
2020] The
paper “VFChain: Enabling Verifiable and Auditable Federated Learning via
Blockchain Systems” has been accepted by IEEE
Transactions on Network Science and Engineering. Congratulations to Dr.
PNEG Zhe! [5 Dec
2020] The
paper “Exploiting Simultaneous Communications to Accelerate Data Parallel
Distributed Deep Learning” has been accepted by IEEE INFOCOM 2021. Congratulations to Dr. SHI Shaohuai! [2 Dec
2020] The
paper “Automated Model Design and Benchmarking of Deep Learning Models for
COVID-19 Detection with Chest CT Scans” has been accepted by AAAI 2021. Congratulations to Mr. HE
Xin, Mr. WANG Shihao and all the co-authors! [23 Nov
2020] The
paper “AutoML: A Survey of the State-of-the-Art” has been accepted by Knowledge-Based Systems.
Congratulations to Mr. HE Xin! Paper available on this link before Jan
29, 2021. [11 Nov
2020] The
paper “Energy-Efficient Inference Service of Transformer-Based Deep Learning
Models on GPUs” has received the Best Paper Award of IEEE GreenCom 2020. Congratulations to WANG Yuxin and Dr. WANG
Qiang! [25 Oct
2020] We are organizing the Special Issue on Interplay
Between Machine Learning and Networking Systems, IEEE Network [CFP].
Submission Deadline: 15 April 2021. [21 Oct
2020] The
paper “A Quantitative Survey of Communication Optimizations in Distributed
Deep Learning” has been accepted by IEEE
Network. Congratulations to Dr. SHI Shaohuai and all the co-authors! [Preprint] [19 June
2020] We are organizing the Special Issue on
Communication-Efficient Distributed Machine Learning, IEEE Transactions on Network Science and Engineering [CFP].
Submission Deadline: 1 Dec 2020. [14 June
2020] The
paper “GPGPU
Performance Estimation with Core and Memory Frequency Scaling” has been
accepted by IEEE TPDS. Congratulations
to WANG Qiang! [2 June
2020] I am serving as a General Chair of IEEE DSS-2020 (IEEE
Conference on Data Science and Systems) [CFP].
Submission Deadline: 1 Sept. 2020. [3 April
2020] We are organizing the Special Section on
Opportunities and Challenges to Integrate AI and Big Data, IEEE Transactions on Industrial
Informatics [CFP].
Submission Deadline: 30 July 2020. [22 March
2020] The
paper “ESetStore:
an Erasure-coded Storage System with Fast Data Recovery” has been
accepted by IEEE TPDS. Congratulations
to LIU Chengjian and WANG Qiang! [13 March
2020] The
paper “FMore: An Incentive Scheme of Multi-dimensional Auction for Federated
Learning in MEC” collaborated with Prof. ZENG Rongfei of Northeastern
University has been accepted by IEEE
ICDCS 2020. [22
January 2020] The
paper “FADNet: A Fast and Accurate Network for Disparity Estimation” has been
accepted by IEEE ICRA 2020.
Congratulations to WANG Qiang, SHI Shaohuai, ZHENG Shizhen, and ZHAO Kaiyong! [15
January 2020] The
paper “Layer-wise Adaptive Gradient Sparsification for Distributed Deep
Learning with Convergence Guarantees” has been accepted by ECAI 2020. Congratulations to SHI
Shaohuai, TANG Zhenheng, WANG Qiang, and ZHAO Kaiyong! [10
December 2019] The
paper “Demystifying Tensor Cores to Optimize Half-Precision Matrix Multiply”
collaborated with Prof. WANG Wei
of HKUST has been accepted by IEEE
IPDPS 2020. [6
December 2019] We
have two papers accepted by IEEE
INFOCOM 2020: “Communication-Efficient Distributed Deep Learning with
Merged Gradient Sparsification on GPUs” and “Joint Access Point Placement and
Power-Channel-Resource-Unit Assignment for 802.11ax-Based Dense WiFi with QoS
Requirements”. Congratulations to SHI Shaohuai, WANG Qiang, and Dr. QIU
Shuwei! [19
November 2019] The
paper “Optimizing Batched Winograd Convolution on GPUs” collaborated with Prof. WANG Wei of HKUST has been
accepted by ACM PPoPP 2020. [10 May
2019] The
paper “A Convergence Analysis of Distributed SGD with Communication-Efficient
Gradient Sparsification” has been accepted by IJCAI 2019. Congratulations to SHI Shaohuai, ZHAO Kaiyong, WANG
Qiang, and TANG Zhenheng! [12 April
2019] The
paper “The Impact of GPU DVFS on the Energy and Performance of Deep Learning:
an Empirical Study” has been accepted by ACM
e-Energy 2019. Congratulations to TANG Zhenheng, WANG Yuxin, and WANG
Qiang! [29 March
2019] The
paper “A Distributed Synchronous SGD Algorithm with Global Top-k
Sparsification for Low Bandwidth Networks” has been accepted by IEEE ICDCS 2019. Congratulations to
SHI Shaohuai and all team members! [30
November 2018] The
paper “MG-WFBP: Efficient Data Communication for Distributed Synchronous SGD
Algorithms” has been accepted by IEEE
INFOCOM 2019. Congratulations to SHI Shaohuai! [12
August 2018] The
paper “Performance Modeling and Evaluation of Distributed Deep Learning
Frameworks on GPUs” has received the Best Paper Award of IEEE DataCom 2018. Congratulations to SHI Shaohuai and WANG
Qiang! [31 July 2018] Our team joins Tencent to break the record of training ImageNet on 2048 Nvidia Tesla P40 GPUs. Read the paper at [arXiv] and media report at [Report by Katyanna Quach]. [May
2018] HKBU
ASC18 team won the First Class Award in the Student Supercomputer
Challenge 2018. Congratulations to our students Ni Ronghao, Wang Shihao, Feng
Zijin, Wang Haixin and Wong Tsz Shing! [Report
by HPCWire]. [23 Dec
2017] The
paper “G-CRS: GPU Accelerated Cauchy Reed-Solomon Coding” has been accepted
by IEEE TPDS. Our source code and
data sets can be found here.
Congratulations to LIU Chengjian! [Sept
2017] The
paper “GPGPU Power Estimation with Core and Memory Frequency Scaling” has
been published by ACM
SIGMETRICS Performance Evaluation Review. Congratulations to WANG Qiang! [18 May
2017] We
have two papers accepted by ACM e-Energy
2017. The paper “EPPMiner: An Extended Benchmark Suite for Energy, Power
and Performance Characterization of Heterogeneous Architecture” is one of the
Best Paper
Candidates. Congratulations to WANG Qiang! [28 April
2017] HKBU
ASC17 team won the First Class Award. Congratulations to our
undergraduate students Wang Shihao (Year 2), Zou Xueyan and Cheong
Chin-wang (Year 3), Wei Wenzhou and Ho Chun-san (Year 4)! [Report
by HPCWire and Report
by HKBU]. [24
February 2017] We have received an NVIDIA GPU Grant. [26
November 2016] The
paper “Energy Efficient Real-time Task Scheduling on CPU-GPU Hybrid Clusters”
has been accepted by IEEE INFOCOM 2017.
Congratulations to MEI Xinxin! My presentation at INFOCOM 2017 received the
"Best-in-Session-Presentation"
award. [September
2016] We
have launched the project of “Benchmarking State-of-the-art
Deep Learning Software Tools”. [22
August 2016] Our
work about an autonomous vehicle public transportation system has been
reported by IEEE
Xplore Innovation Spotlight. [18 June
2016] HKUST CSE Alumni Homecoming
Workshop 2016 on Big Data and Deep Learning was successfully held. I gave
a presentation of A tale of two
cities: GPU computing and machine learning, and you can download
the PowerPoint slides here. [18-22
April 2016] HKBU
ASC16 team won the First Class Award and
Most Popular
Team Award. Congratulations to our students Xu Pengfei, Ting
Hoshing, Cheng Guanlun, Du Jiangyang and Ho Chun-san! [Report
by HPCWire and Report by HKBU]. [26 March
2016] The
paper "Dissecting GPU Memory Hierarchy through Microbenchmarking"
has been accepted by IEEE TPDS. Our
source code and data sets can be found here.
Congratulations to Mei Xinxin! [1 August
2015] The
paper "R-Memcached: a Reliable In-Memory Cache System for Big Key-Value
Stores" has received the Best Paper Award of
BigCom 2015. Congratulations to Liu
Chengjian! [24 March
2015] The
paper "Community-based Bus System as Routing Backbone for Vehicular Ad
Hoc Networks" has been accepted by
ICDCS 2015. [17 Nov
2014] The
paper "Online Procurement Auctions for Resource Pooling in
Client-Assisted Cloud Storage Systems" has been accepted by IEEE INFOCOM 2015. [23 April
2014] The
paper "Accelerating the Scoring Module of Mass Spectrometry-Based
Protein Identification Using GPUs" has been accepted by BMC Bioinformatics. Congratulations to
Li You! [23 Feb
2014] The
paper "Dissecting Darknets: Measurement and Performance Analysis"
has been accepted by ACM Transactions
on Internet Technology. Congratulations to Chen Xiaowei! [25 Jan
2014] The
paper "G-BLASTN: Accelerating Nucleotide Alignment by Graphics
Processors" has been published online in Bioinformatics. Congratulations to Zhao Kaiyong! [22 Aug
2013] G-BLASTN 1.0 Released! G-BLASTN is a GPU-accelerated nucleotide alignment
tool based on the widely used NCBI-BLAST. It can produce exactly the same
results as NCBI-BLAST. In comparison to the multithreaded NCBI-BLAST
on Intel Core i7-3820 (quad-core, 3.6GHz), G-BLASTN can achieve an
average of 7X
speedup for masked human and mouse
genome databases using a single NVIDIA GTX780 card. |