江家浩 (Chiang Chia-Hao)

M.S. at CSIE, NTNU
B.S. at MS, NTU

  Hsinchu, Hsinchu City, Taiwan

    


工作經歷

八月 2023 - Present

研究助理   中央研究院 Academia Sinica 

1. Design and construct the multi-variable climate downscaling models via machine learning, deep learning methods for precipitation predictions.
2. Create an integrated model for difference climatological simulation models for less computational costs and increasing prediction resolution.

九月 2021 - 二月 2022

EMI助教  國立臺灣師範大學 National Taiwan Normal University

1. English Medium Instruction (EMI) of course content in English.
2. Teaching assistant (TA) of Computer Graphics and Data Visualization.
3. Help undergraduate students solving problems.
4. Correct assignment.

八月 2018 - 六月 2021

輔導老師  新竹市私立陳杰文理短期補習班

1. Help and teaching high-school students to solve problems of all-subjects.
2. Teaching and delivering methodology of studying.
3. Arranging answer elaboration of questions of math competitions.

五月 2019 - 六月 2020

工程師  頎邦科技股份有限公司 (Chipbond Technology Inc.)

1. Increase the capabilities of manufacturing machines and the flows, designs of producing related systems, cooperating with IT, IE and Production Dept.
2. Increase the yield rate of packaging of driver-IC.
3. Analyze the manufacturing data for yield rate issue. 4. Increase the durability and stability of both by establishing SOP, also the TA (OP) orientation.

二月 2017 - 六月 2018

行政助理工讀生  國立台灣大學 National Taiwan University

協助會計公文的遞送、歸檔,以及報帳、採購案流程

學歷

2021 - 2023

國立臺灣師範大學 (National Taiwan Normal University, NTNU)

資訊工程學系 (Computer Science and Informative Technology)

2014 - 2018

國立臺灣大學 (National Taiwan University, NTU)

材料科學工程學系 (Material Science and Engineering)

資格認證



TOEIC

ETS

六月 2025 到期

JLPT

N3


專案

Deep Learning Based Climate Downscaling Model

We proposed a deep convolutional neural network with skip connections, attention blocks, and auxiliary data concatenation, in order to downscale the low-resolution precipitation of the climate models into high-resolution one. 
We compare with other models and show better performance in metrics of Pearson Correction, MAE, RMSE, SSIM, and quantitative precipitation forecast (QPF).