Gary S. W. Goh | curriculum vitae

Profile

Name:	Gary Goh Shing Wee
Experience:	3 years academic experience in machine learning and predictive modeling + 1 year industrial experience in building a complex recommendation system for an online video live-streaming and community platform
Interest:	Data-driven Artificial Intelligence applications
Proficiency:	Large-scale data processing using Hadoop and Spark SQL and training deep neural networks with deep learning frameworks such as TensorFlow and PyTorch
:	`garygsw@gmail.com`
:	`linkedin.com/in/garygsw`

Work Experience

2021-2022
Data Scientist

Sea Ltd — Garena
- Develop machine learning recall and ranking models for BOOYAH!, an online video live-streaming and community platform, to recommend highly relevant and personalized content to users
- Achieved 50-80% increase in online metrics such as average video duration per user and click-through rates
- Built samples dataset generation pipeline and design features for deep learning models
2014-2016
Research Engineer

Agency for Science, Technology and Research (A*STAR) Singapore Institute of Manufacturing Technology (SIMTech)
- Research in data mining for supply chain and manufacturing systems
- Develop full-stack software for Manufacturing Control Tower program
- Provide text-mining solutions for the Antuit-SIMTech Supply Chain Analytics Lab
- Design and implement full-stack desktop application with data visualization and demand forecasting capability in a T-Up secondment program to JR Vending
2018-2019
Electrify

IoT Software Engineer (Freelance)
- Develop and maintain software for unix-based embedded systems in remote devices to collect sensor data under intermittent internet connection environment
- Utilize MQQT messaging protocol with AWS IoT Framework to ensure reliable and secure data collection
- Create AWS Lambda functions to relay data exchange between content management server to remote devices
2013-2014
Integrated Decision Systems Consultancy (IDSC)

Associate Consultant (Intern)
- Built optimization models for stock mix optimization (OPTIMAX), supply chain network optimization, and inventory optimization for the consultancy's clientele
- Perform simulation analysis to evaluate various inventory re-order policies
- Develop demand forecasting and data visualization capability into software solutions

Education

2016-2021
Singapore University of Technology and Design (SUTD) — ISTD Pillar

Master of Engineering (Research) in Computer Science and Design
- Advised by Prof. Alexander Binder and previously Prof. Yue Zhang
- Cumulative Grade Point Average: 5.00 / 5.00
- Thesis: Analyzing Deep Learning Models for Traffic Prediction
- Thesis committee: Zhou Jianying, Lim Kwan Hui, Berrak Sisman, Liu Jun
- Coursework: Analysis of Algorithms, Statistics, Computer Networks, Theory and Application of Software Security, Database Systems
- Teaching assistant for Computation Structures labs taught by Dr. Oka Kurniawan
- Funded by Singapore-MIT Alliance for Research and Technology (SMART) Fellowship
2010-2014
National University of Singapore (NUS) — Faculty of Engineering

Bachelor of Engineering in Industrial and Systems Engineering
- Area of specialization: Logistics and Systems Optimization
- Cumulative Aggregate Point: 4.32 / 5.00 (2^nd Upper Honors)
- Coursework: Operation Research, Simulation, Statistics, Probability, Data Structures and Algorithms,Software Engineering, Supply Chain and Inventory Management
- Thesis: Multi-mode Switching Problem for Production/Inventory Systems
- Thesis supervised by Prof. Shuangchi He and Prof. Dacheng Yao
- Awards: Faculty’s Dean List (Top 5% of Cohort) – 2011/12 & 2013/14 Semester 2

Selected Publications

ICPR
Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution

Goh, Gary S. W., Lapuschkin, Sebastian, Weber, Leander, Samek, Wojciech, and Binder, Alexander

In 2020 25th International Conference on Pattern Recognition (ICPR)

Abstract arXiv HTML PDF Code Poster Slides Talk Bib

Integrated Gradients as an attribution method for deep neural network models offers simple implementability. However, it suffers from noisiness of explanations which affects the ease of interpretability. The SmoothGrad technique is proposed to solve the noisiness issue and smoothen the attribution maps of any gradient-based attribution method. In this paper, we present SmoothTaylor as a novel theoretical concept bridging Integrated Gradients and SmoothGrad, from the Taylor’s theorem perspective. We apply the methods to the image classification problem, using the ILSVRC2012 ImageNet object recognition dataset, and a couple of pretrained image models to generate attribution maps. These attribution maps are empirically evaluated using quantitative measures for sensitivity and noise level. We further propose adaptive noising to optimize for the noise scale hyperparameter value. From our experiments, we find that the SmoothTaylor approach together with adaptive noising is able to generate better quality saliency maps with lesser noise and higher sensitivity to the relevant points in the input space as compared to Integrated Gradients.
@inproceedings{GohLWSB21Understanding, author = {Goh, Gary S. W. and Lapuschkin, Sebastian and Weber, Leander and Samek, Wojciech and Binder, Alexander}, title = {Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution}, booktitle = {2020 25th International Conference on Pattern Recognition (ICPR)}, pages = {4949--4956}, publisher = {IEEE}, year = {2021}, address = {Virtual Event / Milan, Italy}, doi = {10.1109/ICPR48806.2021.9413242}, arxiv = {2004.10484} }
ICDM
Twitter-Informed Crowd Flow Prediction

Goh, Gary S. W., Koh, Jing Yu, and Zhang, Yue

In 2018 IEEE International Conference on Data Mining Workshops (ICDM)

Abstract HTML PDF Code Poster Slides Bib
Award: Research Showcase Award (Data Science) at FIRST Industry Workshop 2018

We explore the usage of real-time tweets to explain for non-recurring large-scale spatio-temporal crowd movement. The aim is to evaluate the usefulness of tweets to improve the performance of city-wide crowd flow prediction. We conduct experiments in the context of Singapore city to investigate our proposition by extending upon an existing crowd flow prediction model. Implemented using a deep-neural-network-based approach, an end-to-end predictive model is configured to take in tweets as additional inputs to forecast the future flow of crowds in an urban environment. We extract various features from tweets, such as tweet counts, tweet tenses and sentiments as additional signals to the predictive model. From the experimental results, we show that some models are able to improve the prediction accuracy, and share our insights on how tweets are related to crowd flows.
@inproceedings{GohKZ18Twitter, author = {Goh, Gary S. W. and Koh, Jing Yu and Zhang, Yue}, title = {Twitter-Informed Crowd Flow Prediction}, booktitle = {2018 IEEE International Conference on Data Mining Workshops (ICDM)}, pages = {624--631}, publisher = {IEEE}, year = {2018}, address = {Singapore}, doi = {10.1109/ICDMW.2018.00097} }

Note: For the complete publications list, please see the publications page or my Google scholar.

Technical Skills

GitHub:	`https://github.com/garygsw`
Programming languages:	Python, C/C++/C#, Javascript, Java, Microsoft VBA, R
Operating systems:	Mac OS, Windows, Ubuntu, Raspberry Pi OS, Linux Mint
Python libraries:	PyTorch, NumPy, Pandas, SciPy, Keras, Theano, scikit-learn, SpaCy, Django, ...
Frameworks:	Jupyter, Qt, AWS, Miscrosoft .NET 4.5, Cordova
Database systems:	MySQL, PostgreSQL, SQLite, MongoDB
Client-side technologies:	HTML5, CSS3, JQuery, D3.js, AngularJS
Other software:	Automod, AIMMS, Adobe Photoshop and Illustrator

Others

Languages: English (fluent), Chinese (mother tongue)
Hobbies: traveling, computer games, hiking, movies, cooking, photography, digital design, guitar

Profile

Work Experience

Data Scientist

Sea Ltd — Garena

Research Engineer

Agency for Science, Technology and Research (A*STAR) Singapore Institute of Manufacturing Technology (SIMTech)

Electrify

IoT Software Engineer (Freelance)

Integrated Decision Systems Consultancy (IDSC)

Associate Consultant (Intern)

Education

Singapore University of Technology and Design (SUTD) — ISTD Pillar

Master of Engineering (Research) in Computer Science and Design

National University of Singapore (NUS) — Faculty of Engineering

Bachelor of Engineering in Industrial and Systems Engineering

Selected Publications

Technical Skills

Others

Agency for Science, Technology and Research (ASTAR) Singapore Institute of Manufacturing Technology (SIMTech)*