Boyo Chen
Ph.D. student
University of Tokyo
chen@ms.k.u-tokyo.ac.jp

CV

Boyo is a Ph.D. student in University of Tokyo working with Naoto Yokoya. He broadly studies foundational topics in machine learning. His previous researches mainly focus on building up a reliable system for disaster management using computer vision techniques. He received his BS and MS in Computer Science from National Taiwan University in 2016 and 2018, respectively, advised by Hsuan-Tien Lin. Before leaving for ph.D., in 2019, he worked at Appier for about one year to acquire experiences in the industry, where he developed a data-intensive API server, handling large-scale data with graphQL.

Education

April 2021 - present Ph.D. in Complexity Science and Engineering
The University of Tokyo, Tokyo, Japan
September 2016 - July 2018 M.S. in Computer Science and Information Engineering
National Taiwan University, Taipei, Taiwan
September 2012 - July 2016 B.S. in Computer Science and Information Engineering
National Taiwan University, Taipei, Taiwan

Working Experience

December 2019 - March 2021 National Taiwan University, reasearch assistant
January 2019 - November 2019 Appier, Taipei, Taiwan, Software Engineer
September 2018 - December 2018 Military Service in Taiwan.
July 2017 - August 2017 Yahoo, Taipei, Taiwan, Software Engineer Intern

Publications

Rotation-blended CNNs on a new open dataset for tropical cyclone image-to-intensity regression
Boyo Chen, Buo-Fu Chen, Hsuan-Tien Lin
KDD 2018
[abs] [pdf]
In this work, we apply specialized CNN to tropical cyclone intensity regression tasks, set up a remarkable benchmark for disaster management of tropical cyclones. The proposed framework is used for forecasting in Central Weather Bureau, Taiwan.
Estimating Tropical Cyclone Intensity by Satellite Imagery Utilizing Convolutional Neural Networks
Buo-Fu Chen, Boyo Chen, Hsuan-Tien Lin, Russell L. Elsberry
Weather and Forecasting 34 (2), 447-465
[abs] [pdf]
In this journal work, we verified and concluded our work in 2018. The report was published by Weather and Forecasting, the best journal in the related field.
Real-time Tropical Cyclone Intensity Estimation by Handling Temporally Heterogeneous Satellite Data
Boyo Chen, Buo-Fu Chen, Yun-Nung Chen
AAAI 2021
[abs] [pdf]
Due to the limitation of satellites, the previous work of estimating Tropical Cyclone(TC) intensity can only obtain data once every 3 hours. In this work, we use Generative Adversarial Networks to handle missing data. As a result, the frequency for estimating the TC intensity improved from once per 3 hours to once per 15 minutes.
CNN Profiler on Polar Coordinate Images for Tropical Cyclone Structure Analysis
Boyo Chen, Buo-Fu Chen, Chun-Min Hsiao
AAAI 2021
[abs] [pdf]
Besides the intensity, the size of a Tropical Cyclone(TC) is another critical factor for disaster management. We aim to set up a benchmark for objectively and routinely analyzing the TC structure profile, which comprises both intensity and size information. According to a TC's rotational and spiral natures, a specialized convolutional model operating on polar-coordinates, instead of Cartesian coordinates, is proposed.
Accurate and Clear Quantitative Precipitation Nowcasting based on a Deep Learning Model with Consecutive Attention and Rainmap Discrimination
Ashesh, Chia-Tung Chang, Buo-Fu Chen, Hsuan-Tien Lin, Boyo Chen, Treng-Shi Huang
AIES 2022
[abs] [pdf]
We propose a new deep learning model for precipitation nowcasting that includes both the discrimination and attention techniques. The model is examined on a newly-built benchmark dataset that contains both radar data and actual rain data.

Teaching Experience

2016 fall Machine Learning Foundations, TA
About 100 students, course link
2017 spring Machine Learning Techniques, TA
About 130 students, course link
2017 fall Machine Learning Foundations, TA
About 260 students, course link
2018 spring Machine Learning Techniques, TA
About 200 students, course link

Skills

Proficient Python programming.
Machine learning algorithms and applications.
Data-intensive applications.
Intermediate C/C++ programing.
Document-oriented databases, such as MongDB.
Build and maintain Restful API and GraphQL API.
Distributed computing with Apache Spark.
With Primary Knowledge CICD: Jenkins, Docker, Kubernates.
Relation-oriented databases such as PostgreSQL and MySQL.
Object-oriented programming using Java, Android.

Life

I married my wife on July 27, 2018.
My daughter was born on December 6, 2020.
We have two cats, ChiChi and Bolo.

Forked from Brandon Amos