Biography

Chen-Yu started his Ph.D. in fall, 2018 at KAUST.

Combining his interests in fundamental systems and the trend of machine learning, Chen-Yu is collaborating with colleagues on developing efficient distributed machine learning systems, to be specific, trying to alleviate network bandwidth bottleneck by offloading aggregation operations to network devices (see DAIET).

During his time at Academia Sinica, Taiwan, Chen-Yu worked on techniques for digitalizing handwriting and ancient Chinese calligraphy.

Interests

  • Distributed Machine Learning Systems

Education

  • B.S. in Engineering Science and Ocean Engineering, 2016

    National Taiwan University

Recent Publications

Efficient Sparse Collective Communication and its application to Accelerate Distributed Deep Learning

Efficient collective communication is crucial to parallel-computing applications such as distributed training of large-scale …

(* equal contribution) Jiawei Fei*, Chen-Yu Ho*, Atal Narayan Sahu, Marco Canini, Amedeo Sapio
PDF Project

Compressed Communication for Distributed Deep Learning: Survey and Quantitative Evaluation

Powerful computer clusters are used nowadays to train complex deep neural networks (DNN) on large datasets. Distributed training …

Hang Xu, Chen-Yu Ho, Ahmed M. Abdelmoniem, Aritra Dutta, El Houcine Bergou, Konstantinos Karatsenidis, Marco Canini, Panos Kalnis
PDF Code Project

On the Discrepancy between the Theoretical Analysis and Practical Implementations of Compressed Communication for Distributed Deep Learning

Compressed communication, in the form of sparsification or quantization of stochastic gradients, is employed to reduce communication …

Aritra Dutta, El Houcine Bergou, Ahmed M. Abdelmoniem, Chen-Yu Ho, Atal Narayan Sahu, Marco Canini, Panos Kalnis
PDF

Natural Compression for Distributed Deep Learning

Due to their hunger for big data, modern deep learning models are trained in parallel, often in distributed environments, where …
Samuel Horváth, Chen-Yu Ho, Ľudovít Horváth, Atal Narayan Sahu, Marco Canini, Peter Richtárik
PDF

Scaling Distributed Machine Learning with In-Network Aggregation

Training complex machine learning models in parallel is an increasingly important workload. We accelerate distributed parallel training …
Amedeo Sapio, Marco Canini, Chen-Yu Ho, Jacob Nelson, Panos Kalnis, Changhoon Kim, Arvind Krishnamurthy, Masoud Moshref, Dan R. K. Ports, Peter Richtárik
PDF

Projects

*
All KAUST Academia-Sinica

AR-Navigator

JUNCTIONxKAUST 2018 second prize. Integrate Augmented Reality to indoor navigation.

Daiet

DAIET performs data aggregation along network paths using programmable network devices to alleviate communication bottlenecks in distributed machine learning systems

Shaheen Supercomputer Evaluation

Evaluate different processors architectures and programming environment and to reach the technical specifications provided by the chip manufacturers

Chinese Character Extraction

Extract handwritten Chinese characters from manuscripts

handwriting.js

A simple API for the incredible handwriting recognition of Google IME

Font Embedding

efficiently embed Chinese fonts to webpages

Curriculum Vitae

Last Update: January 2021

Contact

  • chenyu.ho@kaust.edu.sa
  • King Abdullah University of Science and Technology (KAUST), Thuwal, Makkah 23955-6900
  • Enter Building 1 and take the lift to work cell 4409-WS17 on Floor 4
  • Please note that the weekend in Saudi Arabia is Friday and Saturday.

Tag Cloud

Academia-Sinica Algorithms Distributed-Systems Hackathon High-Performance-Computing Image-Processing KAUST Machine-Learning Networked-Systems