Scalable. Interactive. Interpretable.
At Georgia Tech, we innovate scalable, interactive, and interpretable tools that amplify human's ability to understand and interact with billion-scale data and machine learning models. Our current research thrusts: human-centered AI (interpretable, fair, safe AI; adversarial ML); large graph visualization and mining; cybersecurity; and social good (health, energy).
Current Polo Club members and recent alum. See all alum.
Nilaksh
CSE PhD
Haekyu
CS PhD
Jay
ML PhD
Austin
ML PhD
Rahul
CS PhD
Seongmin
CS PhD
Ben
ML PhD
Anthony
CS PhD
Sivapriya
Analytics MS
Omar
CS Undergrad
Anish
CS Undergrad
Frank
CS Undergrad
Jon
CS BS/MS
Megan
CS Undergrad
Alex
CS BS/MS
Kevin
CS Undergrad
Pratham
CS Undergrad
Polo
Associate Prof
Scott
Senior Applied Scientist, Microsoft
Minsuk
Asst Prof, Oregon State Univ
Shang
Asst Prof, National Taiwan Univ
Fred
Research Scientist, Apple
Anmol
Data Scientist, The Home Depot
Bob
Software Engineer, Google
Jonathan
Alex
HCI PhD, Carnegie Mellon
Will
HCI PhD, Carnegie Mellon
Robert
Software Engineer, Google
Sasha
CS Undergrad, Fayetteville State University

Human-Centered AI

Interpretable, fair, and safe artificial intelligence, through interactive intelligent visualization, with application in adversarial machine learning (how protect AI from harm, and from doing harm).

VisCUIT: Visual Auditor for Bias in CNN Image Classifier
VisCUIT
Visual Auditor for Bias in CNN Image Classifier
  • CVPR'22
MisVis: Explaining Web Misinformation Connections via Visual Summary
MisVis
Explaining Web Misinformation Connections via Visual Summary
AVAST Software
  • CHI'22
: Python toolbox to study graph vulnerability and robustness
Python toolbox to study graph vulnerability and robustness
Featured in Nvidia Data Science Toolkit
  • CIKM'21
MalNet-Graph: Largest graph representation learning database
MalNet-Graph
Largest graph representation learning database
Oral Presentation
Facebook AI, Microsoft
  • NeurIPS'21
Autograde-Viz: Automatic Grading of D3.js Visualizations
Autograde-Viz
Automatic Grading of D3.js Visualizations
VIS Best Poster
Deployed
  • VIS'21
Argo Scholar: Interactive Visual Exploration of Literature in Browser
Argo Scholar
Interactive Visual Exploration of Literature in Browser
VIS Best Poster, Honorable Mention
Deployed
  • VIS'21
TrafficVis: Fighting Human Trafficking through Visualization
TrafficVis
Fighting Human Trafficking through Visualization
VIS Best Poster, Honorable Mention
  • VIS'21
NeuroCartography: Scalable Automatic Visual Summarization of Concepts in Deep Neural Networks
NeuroCartography
Scalable Automatic Visual Summarization of Concepts in Deep Neural Networks
Top VIS paper invited to SIGGRAPH
  • VIS'21
Dodrio: Exploring Transformer Models with Interactive Visualization
Dodrio
Exploring Transformer Models with Interactive Visualization
  • ACL'21
Accent Pretraining: Robust Accented Speech Recognition with Adversarial Transfer Learning
Accent Pretraining
Robust Accented Speech Recognition with Adversarial Transfer Learning
Amazon AWS AI
  • INTERSPEECH'21
CardiacAR: Cardiovascular Surgical Planning via Augmented Reality
CardiacAR
Cardiovascular Surgical Planning via Augmented Reality
Children's Healthcare of Atlanta
  • Asian CHI'21
EnergyVis: Interactively Tracking and Exploring Energy Consumption for ML Models
EnergyVis
Interactively Tracking and Exploring Energy Consumption for ML Models
  • CHI'21
Human-in-the-Loop NLP: Putting Humans in the Natural Language Processing Loop: A Survey
Human-in-the-Loop NLP
Putting Humans in the Natural Language Processing Loop: A Survey
  • HCI+NLP'21
MalNet-Image: Largest binary image database to date
MalNet-Image
Largest binary image database to date
  • arXiv'21
SkeletonVis: Understanding Adversarial Attacks on Human Action Recognition Models
SkeletonVis
Understanding Adversarial Attacks on Human Action Recognition Models
Intel
  • AAAI'21
AI Guidelines: Survey of Industry AI Design Guidleines
AI Guidelines
Survey of Industry AI Design Guidleines
  • TREX@VIS'20
UnMask: Protecting deep learning through robust feature alignment
UnMask
Protecting deep learning through robust feature alignment
  • Big Data'20
Communicating with Interactive Articles: Examining interactive article design by synthesizing theory from education, journalism, and visualization
Communicating with Interactive Articles
Examining interactive article design by synthesizing theory from education, journalism, and visualization
  • Distill'20
CNN Explainer: Learning Convolutional Neural Networks with Interactive Visualization
CNN Explainer
Learning Convolutional Neural Networks with Interactive Visualization
Top VIS paper invited to SIGGRAPH
Went Viral
  • VAST'20
GOGGLES: Automatic Image Labeling with Affinity Coding
GOGGLES
Automatic Image Labeling with Affinity Coding
  • SIGMOD'20
Bluff: Interactive Interpretation of Adversarial Attacks on Deep Learning
Bluff
Interactive Interpretation of Adversarial Attacks on Deep Learning
  • VIS'20
Chameleon: Understanding and Visualizing Data Iteration in Machine Learning
Chameleon
Understanding and Visualizing Data Iteration in Machine Learning
Apple
  • CHI'20
REST: Robust and Efficient Neural Networks for Sleep Monitoring in the Wild
REST
Robust and Efficient Neural Networks for Sleep Monitoring in the Wild
  • WWW'20
CNN 101: Interactive Visual Learning for Convolutional Neural Networks
CNN 101
Interactive Visual Learning for Convolutional Neural Networks
  • CHI'20
RECAST: Interactive Auditing of Automatic Toxicity Detection Models
RECAST
Interactive Auditing of Automatic Toxicity Detection Models
  • CSCW'21
Massif: Interactive Interpretation of Adversarial Attacks on Deep Learning
Massif
Interactive Interpretation of Adversarial Attacks on Deep Learning
  • CHI'20
Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations
Summit
Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations
  • VAST'19
FairVis: Visual Analytics for Discovering Intersectional Bias in Machine Learning
FairVis
Visual Analytics for Discovering Intersectional Bias in Machine Learning
  • VAST'19
Gamut: Understanding How Data Scientists Understand Machine Learning Models
Gamut
Understanding How Data Scientists Understand Machine Learning Models
Microsoft Research
  • CHI'19
PeopleMap: Mapping Out Researchers using Natural Language Processing
PeopleMap
Mapping Out Researchers using Natural Language Processing
VAST Best Poster Research, Honorable Mention
  • VIS'20
Magic Crop: Cropping Headshot Photo with AI
Magic Crop
Cropping Headshot Photo with AI
ActiVis: Visual Exploration of Facebook Deep Neural Network Models
ActiVis
Visual Exploration of Facebook Deep Neural Network Models
Top VIS paper invited to SIGGRAPH
Went Viral
Deployed
Facebook
  • VAST'17
GAN Lab: Playing with Generative Adversarial Networks in Browser
GAN Lab
Playing with Generative Adversarial Networks in Browser
Went Viral
Google
  • VAST'18
Visual Analytics in Deep Learning: An Interrogative Survey for the Next Frontiers
Visual Analytics in Deep Learning
An Interrogative Survey for the Next Frontiers
  • TVCG
TeleGam: Combining Visualization and Verbalization for Interpretable Machine Learning
TeleGam
Combining Visualization and Verbalization for Interpretable Machine Learning
Microsoft Research
  • VIS'19
NeuralDivergence: Exploring and Understanding Neural Networks by Comparing Activation Distributions
NeuralDivergence
Exploring and Understanding Neural Networks by Comparing Activation Distributions
  • PacificVis'19
Discovering Intersectional Bias: Discovery of Intersectional Bias in Machine Learning Using Automatic Subgroup Generation
Discovering Intersectional Bias
Discovery of Intersectional Bias in Machine Learning Using Automatic Subgroup Generation
  • ICLR'19
SHIELD: Fast, practical defense for deep learning
SHIELD
Fast, practical defense for deep learning
Audience Appreciation Award, Runner-up
Intel
  • KDD'18
ShapeShifter: 1st Targeted Physical Attack on Faster R-CNN Object Detector
ShapeShifter
1st Targeted Physical Attack on Faster R-CNN Object Detector
DARPA Spotlight
Intel
  • PKDD'18
ShapeShop: Understand Deep Learning via Interactive Experimentation
ShapeShop
Understand Deep Learning via Interactive Experimentation
PNNL
  • CHI'17
The Beginner's Guide to Dimensionality Reduction:
The Beginner's Guide to Dimensionality Reduction
Best Paper, Honorable Mention
  • VISxAI'18
Protect AI with JPEG: Simple, powerful defense for deep learning
Protect AI with JPEG
Simple, powerful defense for deep learning
Intel
  • arXiv
Interactive Classification: For Deep Learning Interpretation
Interactive Classification
For Deep Learning Interpretation
  • CVPR'18
ADAGIO: Interactive experimentation with adversarial attack and defense for audio
ADAGIO
Interactive experimentation with adversarial attack and defense for audio
Intel
  • arXiv
ML Cube: Visual Exploration of Machine Learning Results Using Data Cube Analysis
ML Cube
Visual Exploration of Machine Learning Results Using Data Cube Analysis
Facebook
  • HILDA'16
Mixed Reality: MR for Learning Programming
Mixed Reality
MR for Learning Programming
Best Work-in-Progress, Honorable Mention
  • IDC'19
Augmenting Coding: AR for Learning Programming
Augmenting Coding
AR for Learning Programming
Best Poster
  • ChineseCHI'18

Large Graph Exploration & Visualization

Argo Lite: Interactive Graph Visualization in Browser
Argo Lite
Interactive Graph Visualization in Browser
Used by 6000+ students
Deployed
  • CIKM'20
Atlas: Local Graph Exploration in a Global Context
Atlas
Local Graph Exploration in a Global Context
  • IUI'19
VIGOR: Interactive Visual Exploration of Graph Query Results
VIGOR
Interactive Visual Exploration of Graph Query Results
Symantec
  • VAST'17
Graph Playground: Graph Decompositions in 3D
Graph Playground
Graph Decompositions in 3D
  • VAST'17
Carina: Million-scale Graph Visualization in Web Browsers
Carina
Million-scale Graph Visualization in Web Browsers
  • WWW'17
VISAGE (demo): Interactive Visual Graph Querying
VISAGE (demo)
Interactive Visual Graph Querying
Best Demo, Runner-up
  • SIGMOD'17
FACETS: Adaptive graph exploration
FACETS
Adaptive graph exploration
  • SDM'17
ETable: Interactive Browsing and Navigation in Relational Databases
ETable
Interactive Browsing and Navigation in Relational Databases
  • VLDB'16
VISAGE: Interactive Visual Graph Querying
VISAGE
Interactive Visual Graph Querying
  • AVI'16
Scalable Graph Exploration & Visualization: A survey
Scalable Graph Exploration & Visualization
A survey
  • BigComp'15
TourViz: Interactive Visualization of Connection Pathways in Large Graphs
TourViz
Interactive Visualization of Connection Pathways in Large Graphs
  • KDD'12
Apolo: Explore million-node graphs in real time
Apolo
Explore million-node graphs in real time
  • CHI'11
GLO-STIX: Graph-Level Operations for Specifying Techniques and Interactive eXploration
GLO-STIX
Graph-Level Operations for Specifying Techniques and Interactive eXploration
  • InfoVis'14
BubbleNet: Multi-resolution Large Graph Exploration
BubbleNet
Multi-resolution Large Graph Exploration
  • ICDM'13
NaturalMotion: Exploring Gesture Controls for Visualizing Time-Evolving Graphs
NaturalMotion
Exploring Gesture Controls for Visualizing Time-Evolving Graphs
  • VIS'16

Scalable Graph Mining

AttriPart: Local Partition in Rich Graphs
AttriPart
Local Partition in Rich Graphs
  • Big Data'18
Scalable K-Core Decomposition: For Static Graphs Using a Dynamic Graph Data Structure
Scalable K-Core Decomposition
For Static Graphs Using a Dynamic Graph Data Structure
  • Big Data'18
M-Flash: Billion-Scale Graph Computation by Bimodal Block Processing
M-Flash
Billion-Scale Graph Computation by Bimodal Block Processing
  • PKDD'16
M3: billion-scale ML on a PC using virtual memory
M3
billion-scale ML on a PC using virtual memory
  • SIGMOD'16
MMap: Easy billion-scale graph computation on a PC using virtual memory
MMap
Easy billion-scale graph computation on a PC using virtual memory
  • BigData'14
Mobile MMap: Scalable Algorithms on the Go
Mobile MMap
Scalable Algorithms on the Go
  • ScalableML'14
PEGASUS: Billion-Scale Graph Mining on Hadoop
PEGASUS
Billion-Scale Graph Mining on Hadoop
Open Source Software World Challenge, Silver Award
MAGE: Matching Approximate Patterns in Richly-Attributed Graphs
MAGE
Matching Approximate Patterns in Richly-Attributed Graphs

Cyber Security & Fraud

D<sup>2</sup>M: Dynamic Defense and Modeling of Adversarial Movement in Networks
D2M
Dynamic Defense and Modeling of Adversarial Movement in Networks
Microsoft Advanced Threat Protection
  • SDM'20
Cyber MoneyBall: Predicting Cyber Threats with Virtual Security Products
Cyber MoneyBall
Predicting Cyber Threats with Virtual Security Products
Symantec
  • ACSAC'17
FairPlay: Fraud and Malware Detection in Google Play
FairPlay
Fraud and Malware Detection in Google Play
  • SDM'16
MARCO: Fake Review Detection
MARCO
Fake Review Detection
SDM'14 Best Student Paper
  • SDM'14
AESOP: 99%-accurate Malware Detection
AESOP
99%-accurate Malware Detection
Patented
Deployed
Symantec
  • KDD'14
Polonium: Malware Detection over 37 Billion File Relationships
Polonium
Malware Detection over 37 Billion File Relationships
Patented
Deployed
Symantec
  • SDM'11
Latent Gesture: Mobile Protection with Touch Signatures
Latent Gesture
Mobile Protection with Touch Signatures
  • ChineseCHI'14
NetProbe: Auction Fraud Detection
NetProbe
Auction Fraud Detection
  • WWW'07
Quasi Clique: Uncovering Suspicious Links
Quasi Clique
Uncovering Suspicious Links
  • S&P'15
Insider Trading Pattern Discovery:
Insider Trading Pattern Discovery
Securities and Exchange Commission
  • ASONAM'13

Social Good & Health

DeepPop: Deep Learning on Satellite Imagery for Population Estimation
DeepPop
Deep Learning on Satellite Imagery for Population Estimation
Microsoft AI for Earth
  • SIGSPATIAL GEOHUMANITIES'17
Chronodes: Multi-focus mHealth Visual Data Exploration
Chronodes
Multi-focus mHealth Visual Data Exploration
ACM TiiS 2018 Best Paper, Honorable Mention
  • ACM TiiS
mHealth Visual Discovery Dashboard: Making Sense of Mobile Health Data
mHealth Visual Discovery Dashboard
Making Sense of Mobile Health Data
  • Ubicomp'17
Firebird: Predicting Fire Risk in Atlanta
Firebird
Predicting Fire Risk in Atlanta
KDD'16 Best Student Paper, runner-up
Deployed
Atlanta Fire Rescue Department
  • KDD'16
ElectroLens: Understanding Atomistic Simulations
ElectroLens
Understanding Atomistic Simulations
  • VIS'19
PASSAGE
A Travel Safety Assistant with Safe Path Recommendations for Pedestrians
Scalable Architecture in Power Generating Assets: For Anomaly Detection and Visualization
Scalable Architecture in Power Generating Assets
For Anomaly Detection and Visualization
TimeStitch: Interactive Multi-focus Cohort Discovery and Comparison
TimeStitch
Interactive Multi-focus Cohort Discovery and Comparison
Characterizing Smoking and Drinking Abstinence: Using Social Media Communities
Characterizing Smoking and Drinking Abstinence
Using Social Media Communities
Understanding Pediatric Asthma Care: Using Visual Analytics
Understanding Pediatric Asthma Care
Using Visual Analytics
Children's Healthcare of Atlanta

Club Members

Nilaksh Das
CSE PhD
Haekyu Park
CS PhD
Zijie (Jay) Wang
ML PhD
Pokémon trainer.
Austin Wright
ML PhD
football fan and film nut
Rahul Duggal
CS PhD
Seongmin Lee
CS PhD
Ben Hoover
ML PhD
Urban Forager
ShengYun (Anthony) Peng
CS PhD
Sivapriya Vellaichamy
Analytics MS
Omar Shaikh
CS Undergrad
Anish Upadhayay
CS Undergrad
Zhiyan "Frank" Zhou
CS Undergrad
Jon Saad-Falcon
CS BS/MS
Megan Dass
CS Undergrad
Haoyang "Alex" Yang
CS BS/MS
Kevin Li
CS Undergrad
Pratham Mehta
CS Undergrad
Filmmaking
Duen Horng (Polo) Chau
Associate Prof
Covert designer, photographer. Rusty cellist & pianist.

Alumni

Scott Freitas
ML PhD
Senior Applied Scientist, Microsoft
Dr. Acar Tamersoy
CS PhD
Research Scientist, Symantec Research Labs
Dr. Robert Pienta
CSE PhD
Data Science Manager, Intuit
Dr. Chad Stolper
CS PhD
Software Engineer, Google
Dr. Minsuk Kahng
CS PhD
Asst Prof, Oregon State Univ
Human-Centered AI, Data Visualization
Dr. Shang-Tse Chen
CS PhD
Asst Prof, National Taiwan Univ
AI-infused Security: theory + applications for deep learning attack and defense.
Dr. Fred Hohman
CSE PhD
Research Scientist, Apple
Interactive Scalable Interfaces for Machine Learning Interpretability
Jaegul Choo
CSE PhD
Associate Professor, KAIST
Zhiyuan "Jerry" Lin
CS Undergrad
CS PhD, Stanford
Ramen lover and hackathon hacker.
Elias Khalil
CSE Master
Asst Prof. Univ of Toronto
Paras Jain
CS Undergrad
CS PhD, UC Berkeley
Peter Polack
MS Digital Media
PhD Information Studies, UCLA
Samuel Clarke
CS Undergrad
CS Master, Carnegie Mellon
Srishti Gupta
CS Master
Data Engineer, Hike Messenger
Dezhi “Andy” Fang
CS Undergrad
Software Engineer, Citadel
Anmol Chhabria
MS CSE
Data Scientist, The Home Depot
Siwei "Bob" Li
CS Undergrad
Software Engineer, Google
Kristina Marotta
CS PhD, Univ of North Carolina
Matthew Keezer
CS Master
Software Engineer, IBM
Madhuri Shanbhogue
CS Master
Software Engineer ML, Google
Varun Bezzam
CS Master
Software Engineer, Heroku
Prasenjeet Biswal
CS Master
Software Engineer, Yahoo
Nathan Dass
CS Undergrad
Software Engineer, Google
Sam Ford
CS Undergrad
Software Engineer, Square
Joon Kim
CS Undergrad
Site Reliability Engineer, Mailchimp
Jonathan Leo
CS CS/MS
Ángel (Alex) Cabrera
CS Undergrad
HCI PhD, Carnegie Mellon
Will Epperson
CS Undergrad
HCI PhD, Carnegie Mellon
Robert Firstman
CS Undergrad
Guitar player. Music and NBA enthusiast.
Robert Turko
CS Undergrad
Software Engineer, Google
Sasha Richardson
CS Undergrad
CS Undergrad, Fayetteville State University
Dong-Kyu Chae
Postdoc
Asst Prof, Hanyang Univ
Meera Kamath
CS Master
Software Engineer, Microsoft
Florian Foerster
HCI Master
Research Manager, Facebook
Yiqi (Victor) Chen
CS Undergrad
Software Engineer, Facebook
Aakash Goel
CS Master
Google
Shan Li
HCI Master
User Experience Designer, Fortinet
Jake Williams
CS Undergrad
Machine Learning Engineer, Square
Kshitij Kulkarni
ECE Undergrad
EECS PhD, UC Berkeley
Hugo Armando Gualdron Colmenares
CS Master, University of São Paulo
Changhyun Lee
CS Master
Software Enginner, Google
Kaeser Sabrin
CS PhD
Software Engineer, LinkedIn
Sudeep Agarwal
CS Undergrad
Software Engineer, Apple

Data and Visual Analytics

Data and Visual Analytics (DVA) is data science course at Georgia Tech, for both graduate (as CSE6242) and undergraduate students (as CX4242). CSE 6242 is a required core course of the Master of Science in Analytics (MSA). In Spring 2018, the campus version of the course was "mirrored" and offered as a core course of the Online Master of Science in Analytics (OMS Analytics) program. In Fall 2018, the online offering was further expanded to the Online Master of Science in Computer Science (OMS CS) program. Instructors: Dr. Mahdi Roozbahani (since Fall 2019) and Prof. Polo Chau (since Spring 2013).

Year Semester Course Websites Students
2022 Spring Campus 6242 & 4242 Online 6242 1335
2021 Fall Campus 6242 & 4242 Online 6242 1181
2021 Spring Campus 6242 & 4242 Online 6242 1085
2020 Fall Campus 6242 & 4242 Online 6242 1072
2020 Spring Campus 6242 Campus 4242 Online 6242 966
2019 Fall Campus 6242 Campus 4242 Online 6242 877
2019 Spring Campus 6242 & 4242 Online 6242 1000
2018 Fall Campus 6242 & 4242 Online 6242 677
2018 Spring Campus 6242 & 4242 Online 6242 287
2017 Fall Campus 6242 & 4242 273
2017 Spring Campus 6242 & 4242 214
2016 Fall Campus 6242 & 4242 215
2016 Spring Campus 6242 & 4242 187
2015 Fall Campus 6242 & 4242 146
2015 Spring Campus 6242 & 4242 113
2014 Fall Campus 6242 & 4242 118
2014 Spring Campus 6242 & 4242 95
2013 Spring Campus 6242 & 4242 35

KDD IDEA Workshop

Interactive Data Exploration and Analytics. Great IDEAs, 6 years in a row!

Sponsors

Our research has been made possible by support and funding from: NSF (CNS1704701, IIS1563816, IIS1551614, TWC1526254, IIS1217559, CMMI1935548), NIH (1-U54EB020404-01), DARPA (HR00112030001), NASA, Children's Healthcare of Atlanta, Google, Intel, NVIDIA, Amazon, Facebook, Microsoft, Bosch, JPMorgan Chase, Cisco, Symantec, Yahoo, eBay, LogicBlox, LexisNexis Dean's Award, James C. Edenfield Faculty Fellowship, Raytheon Faculty Fellowship.