Kaggle Competitions Expert: https://www.kaggle.com/alanchen1115
(1) Rank No. 5: Aidea AI CUP Mango grade classification, 2020/12/29.
(2) Top 7%: VinBigData Chest X-ray Abnormalities Detection Automatically localize and classify thoracic abnormalities from chest radiographs, 2021/03/30
In this competition, 14 types of thoracic abnormalities will be automatically localized and classified from chest radiographs. The dataset consisting of 18,000 scans have been annotated by experienced radiologists. The model trained with 15,000 independently-labeled images is evaluated on a test set of 3,000 images. These annotations were collected via VinBigData's web-based platform, VinLab.
(3) Top 6%: Understanding Clouds from Satellite Images, Can you classify cloud structures from satellites? 2019/11/19.
There are many ways in which clouds can organize, but the boundaries between different forms of organization are murky. This makes it challenging to build traditional rule-based algorithms to separate cloud features. In this challenge, a keras based Unet model is used to classify cloud organization patterns from satellite images. Techniques used include data augmentation, ensemble modeling, test time augmentation, pseudo label training.
(4) Top 6%: Generative Dog Images, Experiment with creating puppy pics, 2019/08/29.
A generative adversarial network (GAN) is a class of machine learning
system with two neural networks competing with each other in a
game. Traditional GANs generate high-resolution details as a function
of only spatially local points in lower-resolution feature maps. The
self-attention GAN allows attention-driven, long-range dependency
modeling for image generation tasks. In this competition, a tensorflow
based self-attention GAN is used to create images of dogs. There's no
ground truth data to predict. The images submitted to be scored are
based on how well those images are classified as dogs from
pre-trained neural networks.
．08/2019- Data science lecturer for ITRI's AI vocational training class/ AI training lecturer for Taiwan water resources and agriculture research institute.
．08/2001- 07/2019 Assistant professor/head to the department of cosmetic science and application, and department of environmental engineering, Lanyang Institute of Technology.
．01/2000- 07/2001 Vice CEO of the water resources management and policy research center, Tamkang University.
．Ph.D.: School of Civil Engineering, Purdue University (12, 1999). Ph.D. Thesis: “Stochastic Characteristics of Hydrologic and Environmental Time Series”.
．Master Degree: Graduate Institute of Water Resource and Environmental Engineering, Tamkang University (06, 1992). Master Thesis: “Studies on the Non-Gaussian and Non-Stationary Time Series”.
．Bachelor Degree: Department of Chemical Engineering, National Taiwan
University (06, 1990).
Data Scientist / AI Engineer