I'll be joining National Taiwan University as an Assistant Professor in the Department of Computer Science and Information Engineering in February 2020.

I'm looking for motivated students interested in designing secure and trustworthy machine learning models. Please email me if your are interested in joining my research group.

I work in the intersection of applied and theoretical machine learning, with a strong application focus on cybersecurity. My recent research interests include adversarial ML and various aspects of security, privacy, and fairness of ML models.

I received my PhD in computer science from Georgia Tech, advised by Polo Chau and Nina Balcan. During my PhD studies, I co-authored several winning research funding proposals, including a $1.2 million NSF grant (SaTC: CORE: Medium), a $1.5 million gift grant from the Intel Science and Technology Center for Adversary-Resilient Security Analytics, and the IBM PhD Fellowship.


News Oct. 2019Joined Appier as a visiting ML scientist.
Aug. 2019Successfully defended my thesis!
Aug. 2019Presented UnMask at KDD'19 LEMINCS Workshop.
Sept. 2018Presented ShapeShifter and ADAGIO at ECML-PKDD’18.
Aug. 2018Our Shield paper won the KDD’18 Audience Appreciation Award, Runner-Up!

Education

2013 — 2019Ph.D. in Computer Science
Georgia Institute of Technology, Atlanta, GA
Thesis: AI-infused Security: Robust Defense by Bridging Theory and Practice
Committee: Polo Chau (advisor), Nina Balcan (co-advisor), Wenke Lee, Le Song,
                            Kevin Roundy, and Cory Cornelius

2006 — 2010B.S. in Computer Science Information Engineering
National Taiwan University, Taipei, Taiwan

Honors and Awards

2018 — 2019 IBM PhD Fellowship
For my Ph.D. research on “AI-infused Security: Robust Defense by Bridging Theory and Practice” 2018KDD’18 Audience Appreciation Award, Runner Up
For “Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression” 2018NVIDIA GPU Grant
Awarded a NVIDIA Titan V GPU 2016Symantec Fellowship Runner-Up
2016KDD’16 Best Student Paper Award, Runner-Up
For “Firebird: Predicting Fire Risk and Prioritizing Fire Inspections in Atlanta”
2016Amazon Web Services (AWS) in Education Research Grants
$2,500 in AWS cloud credits
2015Amazon Web Services (AWS) in Education Research Grants
$2,500 in AWS cloud credits
2010National Science Council Research Creativity Award
For my undergraduate research on “Link Prediction in Heterogeneous Networks”
2009National Science Council Undergraduate Research Fellowship
A grant that recognizes a small number of outstanding undergraduate researchers
2009KDD Cup 2009 3rd Prize (slow track)
Out of 400+ submissions
KDD CUP is the most prestigious data mining contest

Industry Research Experience

Summer 2018Intel Labs, Hillsboro, OR
Graduate Machine Learning Security Intern
Mentor: Cory Cornelius, Jason Martin
Explored regularization techniques as defense against adversarial attack.

Summer 2017Intel Labs, Hillsboro, OR
Graduate Security Intern
Mentor: Cory Cornelius, Jason Martin
Developed ShapeShifter, the 1st physical adversarial attack that fools Faster R-CNN object detectors

Summer 2016Symantec Research Labs, Culver City, CA
Research Engineer Intern
Mentor: Kevin A. Roundy
Developed patented Virtual Product, a novel framework for enterprise cyber threat detection.

Summer 2015Pindrop Security, Atlanta, GA
Research Intern
Mentor: Raj Bandyopadhyay
Improved phone fraud detection system significantly by 10 absolute percentage.

Academic Research Experience

2013 — 2019Georgia Institute of Technology, Atlanta, GA
Graduate Research Assistant, School of Computational Science and Engineering
Advisors: Polo Chau and Nina Balcan

2011 — 2013Academia Sinica, Taipei, Taiwan
Graduate Research Assistant, Institute of Information Science
Advisors: Chi-Jen Lu and Hsuan-Tien Lin

2008 — 2010National Taiwan University, Taipei, Taiwan
Undergraduate Research Assistant, Department of Computer Science and Information Engineering
Advisors: Shou-De Lin and Hsuan-Tien Lin

Publications

Refereed Conference Papers

ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector
Shang-Tse Chen, Cory Cornelius, Jason Martin, and Duen Horng (Polo) Chau
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD). Dublin, Ireland. Sept. 2018.
Github | Video

Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression
Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Fred Hohman, Siwei Li, Li Chen,
Michael E. Kounavis, and Duen Horng (Polo) Chau
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD). London, UK. Aug. 2018.
Github | Video | Audience Appreciation Award, Runner-Up

Diversified Strategies for Mitigating Adversarial Attacks in Multiagent Systems
Maria-Florina Balcan, Avrim Blum, and Shang-Tse Chen (alphabetic order)
International Conference on Autonomous Agents and Multiagent Systems (AAMAS). Stockholm, Sweden. July 2018.

Predicting Cyber Threats with Virtual Security Products
Shang-Tse Chen, Yufei Han, Duen Horng (Polo) Chau, Christopher Gates, Michael Hart, and
Kevin Roundy
Annual Computer Security Applications Conference (ACSAC). Orlando, FL. Dec. 2017.
Patented

Firebird: Predicting Fire Risk and Prioritizing Fire Inspections in Atlanta
Michael Madaio, Shang-Tse Chen, Oliver Haimson, Wenwen Zhang, Xiang Cheng,
Matthew Hinds-Aldrich, Duen Horng (Polo) Chau, and Bistra Dilkina.
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD). San Francisco, CA. Aug. 2016.
Site | Github | Video | Best Student Paper Award, Runner-Up

Communication Efficient Distributed Agnostic Boosting
Shang-Tse Chen, Maria-Florina Balcan, and Duen Horng (Polo) Chau
International Conference on Artificial Intelligence and Statistics (AISTATS). Cadiz, Spain. May 2016.

Boosting with Online Binary Learners for the Multiclass Bandit Problem
Shang-Tse Chen, Hsuan-Tien Lin, and Chi-Jen Lu
International Conference on Machine Learning (ICML). Beijing, China. June 2014.

An Online Boosting Algorithm with Theoretical Justifications
Shang-Tse Chen, Hsuan-Tien Lin, and Chi-Jen Lu
International Conference on Machine Learning (ICML). Edinburgh, Scotland. June 2012.
Code

Journal Articles and Book Chapters

Chronodes: Interactive Multi-focus Exploration of Event Sequences
Peter J. Polack, Shang-Tse Chen, Minsuk Kahng, Kaya De Barbaro, Rahul Basole, Moushumi Sharmin, Duen Horng (Polo) Chau
ACM Transactions on Interactive Intelligent Systems (TiiS) Special Issue on Interactive Visual Analysis of Human and Crowd Behaviors, 2018.
Video

Exploratory Visual Analytics of Mobile Health Data: Sensemaking Challenges and Opportunities
Peter J. Polack, Moushumi Sharmin, Kaya de Barbaro, Minsuk Kahng, Shang-Tse Chen, and
Duen Horng (Polo) Chau
Mobile Health: Sensors, Analytic Methods, and Applications. Springer, 2017.

Constructing, Analyzing and Visualizing Social Networks: Exemplified by the Academia Social Network in Taiwan
Cheng-Te Li, Chun-Min Chang, Chien-Pang Liu, Shang-Tse Chen, and Shou-De Lin
Journal of Librarianship and Information Studies, 67:72-87, 2008.

Refereed Workshop, Poster, and Demo papers

Extracting Knowledge For Adversarial Detection and Defense in Deep Learning
Scott Freitas, Shang-Tse Chen, and Duen Horng Chau
KDD Workshop on Learning and Mining for Cybersecurity (LEMINCS). Anchorage, AK, Aug. 2019.

Talk Proposal: Towards the Realistic Evaluation of Evasion Attacks using CARLA
Cory Cornelius, Shang-Tse Chen, Jason Martin, and Duen Horng Chau
Dependable and Secure Machine Learning (DSML). Portland, OR, June 2019.

ADAGIO: Interactive Experimentation with Adversarial Attack and Defense for Audio
Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Li Chen, Michael E. Kounavis, and
Duen Horng (Polo) Chau
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD) (demo). Dublin, Ireland. Sept. 2018.
Video

Physical Adversarial Attack on Object Detectors
Shang-Tse Chen, Cory Cornelius, Jason Martin, and Duen Horng (Polo) Chau
ACM KDD Project Showcase. London, UK. Aug. 2018.

Compression to the Rescue: Defending from Adversarial Attacks Across Modalities
Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Fred Hohman, Siwei Li, Li Chen,
Michael E. Kounavis, and Duen Horng (Polo) Chau
ACM KDD Project Showcase. London, UK. Aug. 2018.

TimeStitch: Interactive Multi-focus Cohort Discovery and Comparison
Peter J. Polack, Shang-Tse Chen, Minsuk Kahng, Moushumi Sharmin, and Duen Horng (Polo) Chau
IEEE Conference on Visual Analytics Science and Technology (VAST’15) (Poster). Chicago, IL, Oct. 2015.
Video

Spotting Suspicious Reviews via (Quasi-)clique Extraction
Paras Jain, Shang-Tse Chen, Mozhgan Azimpourkivi, Duen Horng (Polo) Chau, and
Bogdan Carbunar
IEEE Symposium on Security and Privacy (Oakland) (poster). SAN JOSE, CA. May 2015.

An Ensemble of Three Classifiers for KDD Cup 2009: Expanded Linear Model, Heterogeneous Boosting, and Selective Naive Bayes
Hung-Yi Lo, Kai-Wei Chang, Shang-Tse Chen, Tsung-Hsien Chiang, Chun-Sung Ferng, Cho-Jui Hsieh, Yi-Kuang Ko, Tsung-Ting Kuo, Hung-Che Lai, Ken-Yi Lin, Chia-Hsuan Wang, Hsiang-Fu Yu, Chih-Jen Lin, Hsuan-Tien Lin, and
Shou-de Lin
JMLR Workshop and Conference Proceedings, V.7, 57-64, 2009.
3rd Place of the KDD Cup’09 Slow Track

Technical Reports

Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compression
Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Fred Hohman, Li Chen, Michael E. Kounavis, and Duen Horng (Polo) Chau
arXiv:1705.02900, May 2017.

An Ensemble Ranking Solution to the Yahoo! Learning to Rank Challenge
Ming-Feng Tsai, Shang-Tse Chen, Yao-Nan Chen, Chun-Sung Ferng, Chia-Hsuan Wang,
Tzay-Yeu Wen, and Hsuan-Tien Lin
National Taiwan University, Technical Report, Sept. 2010.

Teaching Experience

Teaching Assistant

Fall 2015CSE-6040: Computing for Data Analytics
Instructor: Richard Vuduc
Georgia Institute of Technology, Atlanta, GA
Introductory data analytics course for MS in Analytics students. (37 students)

Spring 2015CS-7545: Machine Learning Theory
Instructor: Santosh Vempala
Georgia Institute of Technology, Atlanta, GA
Advanced ML theory course primarily taken by PhD students. (26 students)

Guest Lectures

Firebird: Predicting Fire Risk and Prioritizing Fire Inspections in Atlanta
Georgia Institute of Technology, Atlanta, GA
CSE-6242 Georgia Tech. Instructor: Polo Chau.
Fall 2018 245 students (32 undergrads)
Spring 2018 200 students (45 undergrads)
Fall 2017 273 students (40 undergrads)

Fall 2015Data analysis and visualization
Georgia Institute of Technology, Atlanta, GA
CSE-6040 Georgia Tech. Instructor: Richard Vuduc. 37 students

Spring 2015Practical ML for Big Data
Georgia Institute of Technology, Atlanta, GA
CS-7545 Georgia Tech. Instructor: Santosh Vempala. 26 students

Course Design

Fall 2016Big Data Bootcamp
Georgia Institute of Technology, Atlanta, GA
Co-led the design of a two-day intense hands-on bootcamp on big data tools, that is now
offered yearly to students in the MS in Analytics program (∼ 50 students each year).
Course material: http://www.sunlab.org/teaching/cse8803/fall2016/lab/

Mentoring

2016 — 2017Madhuri Shanbhogue
M.S. CS, Georgia Tech
Adversarial Machine Learning
Now: Software Engineer at Facebook

2015 - 2016Peter Polack
M.S. CS, Georgia Tech
Interactive visualization of health sensor data
Now: PhD student at UCLA

2014 — 2017Paras Jain
B.S. CS, Georgia Tech
Fraud Detection on Yelp
Now: PhD student at UC Berkeley

Grants and Funding

2018IBM PhD Fellowship
$95,000 over 2 years, covering full Tuition + $35,000 stipend for 2 years

2017SaTC: CORE: Medium: Understanding and Fortifying Machine Learning Based Security Analytics
NSF CNS 1704701
PI: Polo Chau Co-PIs: Taesoo Kim, Wenke Lee, Le Song
Funded: $1,200,000, 8/1/2017 - 7/31/2021
Co-authored winning proposal, contributing a theory-guided decision-making and defense framework

2016Intel Science & Technology Center for Adversary-Resilient Security Analytics (ISTC-ARSA)
PI: Wenke Lee
Co-PIs: Polo Chau, Taesoo Kim, Le Song
Gift Funding: $1,500,000, 2016 - 2019
Co-authored winning proposal, contributing robust, adaptive algorithms for efficient defenses

Invited Talks

Communication Efficient Distributed Agnostic Boosting
Apr. 2016 HotCSE Seminar, Georgia Tech, Atlanta, GA

Boosting with Online Binary Learners for the Multiclass Bandit Problem
June 2014Appier Inc., Taipei, Taiwan.

Press

Oct. 2018“Study reveals new vulnerability in self-driving cars” Tech HQ. Sept. 2018“Erasing Stop Signs: ShapeShifter Shows Self-Driving Cars Can Still Be Manipulated” Georgia Tech, College of Computing. June 2018“Georgia Tech Teams up with Intel to Protect Artificial Intelligence from Malicious Attacks Using SHIELD.” Georgia Tech, College of Computing. Apr. 2018“CSE Ph.D. Students Claim Three Prestigious Fellowships.” Georgia Tech, College of Computing.

Professional Activities

Program Committee

International Conference on Machine Learning (ICML) 2018 - 2019 Annual Conference on Neural Information Processing Systems (NIPS) 2017 - 2019 AAAI Conference on Artificial Intelligence (AAAI) 2018 - 2020 Uncertainty in Artificial Intelligence (UAI) 2015 - 2019 SIAM International Conference on Data Mining (SDM) 2019 Asian Conference on Machine Learning (ACML) 2017 - 2019 Conference on Technologies and Applications of Artificial Intelligence (TAAI) 2014 Deep Learning and Security Workshop @ IEEE S&P (DLS) 2019

Reviewer

International Conference on Artificial Intelligence and Statistics (AISTATS) 2019 Deep Learning and Security Workshop @ IEEE S&P (DLS) 2018 SIAM International Conference on Data Mining (SDM) 2016 - 2017 ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2017
Annual Network and Distributed System Security Symposium (NDSS) 2017 USENIX Security Symposium (USENIX Security) 2017 ACM Conference on Computer and Communications Security (CCS) 2017