Dxlnmqkx3x0l6am9b2kb

YU-SHENG, HUANG 黃宇生

Taipei, TW

0988761120
[email: [email protected]]

Education

National Taiwan University                                                                                                                                            Taipei, Taiwan 

Statistics, Master Degree     GPA 3.94/4.3 (Rank 1th)                                                                                  2016.09 - 2018.06

National Chengchi University                                                                                                                                        Taipei, Taiwan

Statistics, Bachelor Degree                                                                                                                                          2013.09 - 2016.06

Work experience

Vanguard International Semiconductor Co.                                                                                        Hsinchu, Taiwan 

algorithm engineer                                                                                                                                                                 2018.10 - 2020.3
  • Designed an anomaly detection model for monitoring the health of semiconductor manufacturing equipments. 
    • Built and combined three models, Moving Average Model, AutoEncoder and Multi-Scale Convolutional Recurrent Encoder Decoder (MSCRED), to improve higher accuracy. 
    • Used SAS JMP to perform data preprocessing and python with Tensorflow framework to build the model. 
    • Saved every module engineer 1 hour per day. 
  • Designed a wafer defect detection and classification model on photos provided from Automated Optical Inspection (AOI).
    • Used object detection model, Faster R-CNN, with Tensorflow framework. 
    • Achieved 85% accuracy, and 90% recall rate. 
  • Design Automatic Virtual Metrology to update the parameter settings of equipments in real time. 
    • Used Dense Neural Network with Keras framework. 
    • Equipment parameters were estimated and updated 

Publication

Semiparametric regression analysis of current status data under sequential monitoring

  • Developed a semiparametric estimation method for regression analysis based on the sequential monitoring data. 
  • Introduced the additive hazards regression on sequential data to utilize the comprehensive monitoring information.
  •  Proposed a two-stage estimation procedure by pooling the sequence of the current status at monitoring times to estimate the regression coefficients in the semiparametric additive hazard model. 
  • Used R to conduct extensive simulation studies with various censoring rates and monitoring frequencies to investigate the performance. The result indicated that this model has a good performance, which has bias less than 0.01 and is close to the right censored data result. 

Award

The 5th E.SUN commercial bank SAS competition: excellent work

Big Data Data Scientist Competition Text Analysis and Digital Marketing Competition
  • Lead a team of four, including one member majoring marketing, to develop feasible marketing strategy candidates, and then designed further analyzing procedures and models respectively. 
  • Implemented descriptive statistics with SAS Text Miner to analyze forum texts and search logs from the official site of E.SUN, to identify different costumer groups and their corresponding consumption propensities. 
  • Implemented a Decision Tree model, with SAS, SAS VA and SAS Viya, to predict customers’ purchasing power based on customer profile and their credit card history, which achieved 86% accuracy. 
  • Based the analysis and model, we selected the most important variables and decided our major target group, and proposed our final marketing strategy. 

Skills

  • python, R,
  • SAS, SAS JMP
  • pytorch, tensorflow, keras

 selected courses

  • Mathematical statistics (2016 Fall)                                                                                                                              A 
  • Applied Bayesian statistical method (2016 Fall)                                                                                                         A- 
  • Biostatistics research methods (2016 Fall)                                                                                                                 A 
  • Principles and applications of computational biology (2017 Spring)                                                                         A
  •  Machine learning (2017 Spring)                                       A- 
  • Advanced Medical Statistics Method 1 (2017 Fall)                               A+ 
  • Survival analysis (2018 Spring)                                        A+ 
  • Category analysis (2018 Spring)                                       A+