ML@GT Blog

New Ph.D. Program Highlights Growing Importance of Machine Learning

Eight Georgia Tech schools partner to offer advanced degree in emerging field of machine learning. The Georgia Institute of Technology has been approved to offer a new advanced degree program for the emerging field of machine learning.In a unanimous vote, the Board of Regents of the University System of Georgia approved Georgia Tech’s request to establish a Doctor of Philosophy in Machine Learning.“The field of machine learning is now ubiquitous in everything we do. It impacts everything from robotics and cybersecurity to data analytics – all topics of extraordinary interest to Georgia Tech,” said Rafael L. Bras, Georgia Tech provost and executive vice president for Academic Affairs and the K. Harrison Brown Family Chair.“This new Ph.D. program embraces the interdisciplinary impact and nature of machine learning and serves to strengthen Georgia Tech’s strong position as a leading center of knowledge and expertise in this increasingly important field of study.”

Source: New Ph.D. Program Highlights Growing Importance of Machine Learning | College of Computing

Friday, June 16, 2017 - 05:01

ICML 2017 accepted papers and ML@GT

The list of accepted papers at ICML2017 was released yesterday and Andrej Karpathy has published a very nice post breaking down the acceptance by institution. Out of 1701 submissions 433 papers were accepted (or roughly 25.46%) from 420 different institutions. I am excited to see a very strong representation of Machine Learning @ Georgia Tech (ML@GT) with 13 papers (Andrej reported 14 but I could only find 13 — going with the conservative estimate). This brings GT into the Top 10 among institutions and if you only consider academic institutions it is 6th on the list being testament to GT’s strong research in ML. This also nicely ties in with the newly established Ph.D. program in Machine Learning where we will be admitting for fall this year for the first time creating an even more vibrant Machine Learning community at Georgia Tech tightly integrating research and education.

List of accepted ICML 2017 with Georgia Tech affiliation — Congratulations everyone (and of course also to the competition)!

  1. Lazifying Conditional Gradient Algorithms
    Gábor Braun (Georgia Institute of Technology) · Sebastian Pokutta (Georgia Tech) · Daniel Zink (Georgia Institute of Technology)
  2. Conditional Accelerated Lazy Stochastic Gradient Descent
    Guanghui Lan (Georgia Institute of Technology) · Sebastian Pokutta (Georgia Tech) · Yi Zhou (Georgia Institute of Technology) · Daniel Zink (Georgia Institute of Technology)
  3. Learning Hawkes Processes from Short Doubly-Censored Event Sequences
    Hongteng Xu (Georgia Institute of Technology) · Dixin Luo (University of Toronto) · Hongyuan Zha (Georgia Institute of Technology)
  4. Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Prediction
    Wen Sun (Carnegie Mellon University) · Arun Venkatraman (Carnegie Mellon University) · Geoff Gordon (Carnegie Mellon University) · Byron Boots (Georgia Tech) · Drew Bagnell (Carnegie Mellon University)
  5. Survival HMM: An Interpretable, Event-time Prediction Model for mHealth
    Walter Dempsey (University of Michigan) · Alexander Moreno (Georgia Institute of Technology) · Jim Rehg (Georgia Tech) · Susan Murphy (University of Michigan)
  6. Stochastic Generative Hashing
    Bo Dai (Georgia Tech) · Ruiqi Guo (Google Research) · Sanjiv Kumar (Google Research, NY) · Niao He (UIUC) · Le Song (Georgia Institute of Technology)
  7. Prediction under Uncertainty in Sparse Spectrum Gaussian Processes with Applications to Filtering and Control
    Yunpeng Pan (Georgia Tech) · Xinyan Yan (Georgia Institute of Technology) · Evangelos Theodorou (Georgia Tech) · Byron Boots (Georgia Tech)
  8. Online Multiview Learning: Dropping Convexity for Better Efficiency
    Zhehui Chen (Georgia Institute of Technology) · Lin Yang (Johns Hopkins) · Chris Junchi Li (Princeton University) · Tuo Zhao (Georgia Institute of Technology)
  9. Variational Policy for Guiding Point Processes
    Yichen Wang (Gatech) · Grady Williams (Georgia Tech) · Evangelos Theodorou (Georgia Tech) · Le Song (Georgia Institute of Technology)
  10. Fake News Mitigation via Point Process Based Intervention
    Mehrdad Farajtabar (Georgia Tech) · Jiachen Yang (Georgia Institute of Technology) · Xiaojing Ye (Georgia State University) · Huan Xu (Georgia Tech) · Shuang Li () · Rakshit Trivedi (Georgia Institute of Technology) · Elias Khalil (Georgia Tech) · Le Song (Georgia Institute of Technology) · Hongyuan Zha (Georgia Institute of Technology)
  11. Iterative Machine Teaching
    Weiyang Liu (Georgia Tech) · Bo Dai (Georgia Tech) · Jim Regh (Georgia Tech) · Le Song (Georgia Institute of Technology)
  12. Emulating the Expert: Inverse Optimization through Online Learning
    Sebastian Pokutta (Georgia Tech) · Andreas Bärmann (FAU Erlangen-Nürnberg) · Oskar Schneider (FAU Erlangen-Nürnberg)
  13. Know-Evolve: Deep Learning for Temporal Reasoning in Dynamic Knowledge Graphs
    Rakshit Trivedi (Georgia Institute of Technology) · Hajun Dai (Georgia Tech) · Yichen Wang (Gatech) · Le Song (Georgia Institute of Technology)

Friday, May 26, 2017 - 14:06

Meet These Incredible Women Advancing A.I. Research, including our very own Devi Parikh

Our own Devi Parikh was featured as one of the “Incredible Women Advancing A.I. Research” by Forbes. Congrats to Devi.

Devi Parikh is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a Visiting Researcher at Facebook AI Research (FAIR). After receiving her masters and Ph.D. in Electrical and Computer Engineering from Carnegie Mellon, she’s held multiple visiting positions at top research labs and won accolades such as the 2017 IJCAI Computers and Thought award, considered “the premier award for AI researchers under the age of 35”.

Parikh’s most proud of her research work in Visual Question Answering (VQA) which lies at the intersection of computer vision and natural language processing (NLP). “Through making our large datasets and systems publicly available, we’ve enabled research groups around the world to make significant progress on building machines that can automatically answer questions about visual content,” she highlights. Such technology can aid the visually impaired and transmit information on low-bandwidth networks that can’t support images.

Advances in VQA also improve existing product experiences. “We’ll see more and more conversational agents – be it personal assistants or chatbots – that can see, or augmented reality experiences that are visually intelligent.”

Source: Meet These Incredible Women Advancing A.I. Research

Wednesday, May 24, 2017 - 12:34

Home Depot Deep Learning Competition At Georgia Tech

A couple of weeks ago, Home Depot hosted a Deep Learning competition, in partnership with The Agency (the undergraduate ML club) and the Big O Theory club. Here’s a note from the organizers about the event:

Teams of Georgia Tech students spent nearly 24 hours starting the evening of Friday, April 14th, racing to produce the best results on five challenging deep learning problems in a competition hosted by The Home Depot, Big O Theory Club, and The Agency. The problems included time series prediction, image generation, determining the gender of speakers, determining which department a product belongs to based on the product’s image, and determining which search engine would return the best results for different search terms. Fueled by cookies, coffee, energy drinks, the teams produced remarkable results. Two teams completed all three challenges, and the best submissions used cutting-edge loss functions, fine-tuning techniques, data sampling algorithms, and time series analysis techniques to produce incredible results. The winners are as follows:

1st Place: Yuyu Zhang, Hanjun Dai, and Weiyang Liu, PhD students under Prof. Le Song, who each won GTX1080 graphics cards

2nd Place: Raphael Gontijo Lopes, Saurabh Kumar, and Robby Guthrie, undergraduates who each won drones

3rd Place: Shang-Tse Chen, Siddharth Gururani, and Chih-Wei Wu, graduate students in Music Science and Technology who each won Arduino development kits.

We hope to host this competition again this fall, and are looking forward to presenting even more interesting problems and seeing the awesome solutions the Georgia Tech machine learning community creates.

Thursday, May 11, 2017 - 03:06

ML@GT Distinguished Seminar by Pedro Domingo (UW) on “Sum-Product Networks: The Next Generation of Deep Models” 4/19/2017 @ 12n EBB

PedroDomingoTitle: “Sum-Product Networks: The Next Generation of Deep Models”

Speaker: Pedro Domingo (University of Washington)

Date/Time: April 19, 2017, @ 12n (lunch served at 11:30am)

Location: Engineered Biosystems Building (EBB), CHOA Room

Abstract: The two main types of deep learning are function approximation and probability estimation. Function approximators like convolutional neural networks are robust and allow for real-time inference, but are very inflexible, requiring fixed inputs and outputs and detailed supervision. Probability estimators like deep Boltzmann machines allow arbitrary inputs and outputs and require no supervision, but are not robust and do not allow real-time inference.
Both are very opaque. Sum-product networks (SPNs) are a new class of deep models that are suitable for both function approximation and probability estimation. SPNs allow for real-time inference, are robust and comprehensible, and are highly flexible, with any choice of inputs and outputs and any amount of supervision. I will present generative and discriminative algorithms for learning SPN weights, and an algorithm for learning SPN structure. SPNs have achieved impressive results in a wide variety of domains, including object recognition, image completion, activity recognition, language modeling, collaborative filtering, and click prediction, and are arguably the most powerful class of deep models available today. (Joint work with Abe Friesen, Rob Gens, Mathias Niepert and Hoifung Poon.)

Bio: Pedro Domingos is a professor of computer science at the University of Washington and the author of “The Master Algorithm”. He is a winner of the SIGKDD Innovation Award, the highest honor in data science. He is a Fellow of the Association for the Advancement of Artificial Intelligence and has received a Fulbright Scholarship, a Sloan Fellowship, the National Science Foundation’s CAREER Award, and numerous best paper awards. His research spans a wide variety of topics in machine learning, artificial intelligence, and data science, including scaling learning algorithms to big data, maximizing word of mouth in social networks, unifying logic and probability, and deep learning.

Spherical Image

Thursday, April 20, 2017 - 12:03

ML@GT Seminar by Le Song (CSE) on “Embedding as a Tool for Algorithm Design” on April 5, 2017, 12:00n in EBB CHOA Room

SpeakerLe Song,  Computational Science and Engineering (CSE), GA Techls

Location: Engineered Biosystems Building, CHOA Room (Map)

Date/Time: April 5, 2017, 12:00n – 1:00pm (Lunch at 11:30am)

Title: Embedding as a Tool for Algorithm Design

Abstract: Many big data analytics problems are intrinsically complex and hard, making the design of effective and scalable algorithms very challenging. Domain experts need to perform extensive research, and experiment with many trial-and-errors, in order to craft approximation or heuristic schemes that meet the dual goals of effectiveness and scalability. Very often, restricted assumptions about the data, which are likely to be violated in real world, are made in order for the algorithms to work and obtain performance guarantees. Furthermore, previous algorithm design paradigms seldom systematically exploit a common trait of real-world problems: instances of the same type of problem are solved repeatedly on a regular basis, differing only in their data. Is there a better way to design effective and scalable algorithms for big data analytics?

I will present a framework for addressing this challenge based on the idea of embedding algorithm steps into nonlinear spaces, and learn these embedded algorithms from problem instances via either direct supervision or reinforcement learning. In contrast to traditional algorithm design where every step in an algorithm is prescribed by experts, the embedding design will delegate some difficult algorithm choices to nonlinear learning models so as to avoid either large memory requirement, restricted assumptions on the data, or limited design space exploration. I will illustrate the benefit of this new design framework using large scale real world data, including a materials discovery problem, a recommendation problem over dynamic information networks, and a problem of learning combinatorial algorithms over graphs. The learned algorithms can reduce memory usage and runtime by orders of magnitude, and sometimes result in drastic improvement in predictive performance.

Bio: Le Song is an Associate Professor in the Department of Computational Science and Engineering, College of Computing, and an Associate Director of the Center for Machine Learning, Georgia Institute of Technology. He received his Ph.D. in Machine Learning from University of Sydney and NICTA in 2008, and then conducted his post-doctoral research in the Department of Machine Learning, Carnegie Mellon University, between 2008 and 2011. Before he joined Georgia Institute of Technology in 2011, he was a research scientist at Google briefly. His principal research direction is machine learning, especially nonlinear models, such as kernel methods and deep learning, and probabilistic graphical models for large scale and complex problems, arising from artificial intelligence, network analysis, computational biology and other interdisciplinary domains. He is the recipient of the Recsys’16 Deep Learning Workshop Best Paper Award, AISTATS’16 Best Student Paper Award, IPDPS’15 Best Paper Award, NSF CAREER Award’14, NIPS’13 Outstanding Paper Award, and ICML’10 Best Paper Award. He has also served as the area chair for many leading machine learning and AI conferences such as ICML, NIPS, AISTATS, AAAI and IJCAI, and the action editor for JMLR.

Tuesday, April 18, 2017 - 11:26

ML@GT Launch and Welcome Celebration, April 17, 2017 11am-5pm, TSRB Auditorium


The Machine Learning Center at Georgia Tech (ML@GT) invites you to the ML@GT Welcoming Celebration for faculty and students from across campus who are engaged in the field of Machine Learning. Come listen to those in the ML community speak about new research and how ML is changing our world. The day will feature talks from our faculty and a poster session by our students showcasing our best work. 

  • WhenMonday April 17, 2017, 11am-5pm (Reception following) (lunch will be served)
  • Where: Technology Square Research Building (TSRB) Banquet Hall (map)
  • RSVP required HERE (by April 3rd).

Tuesday, April 18, 2017 - 11:26

CyberLaunch an Atlanta-based accelerator for information security and machine learning startups hosting demo day on March 29, 2017 at 3pm

CyberLaunch is hosting it’s Winter 17′ Demo Day to showcase the current portfolio companies on March 29. The event will be held at the Atlanta Tech Village with over 250 investors and innovators, including 20+ cybersecurity and machine learning startups from around the world.

  • Please register for our demo day HERE
  • Location: Atlanta Tech Village – 3423 Piedmont Road NE – Atlanta, Georgia 30305
  • If you are a startup looking to present at the startup showcase, apply HERE
  • Date/Time: Wednesday, March 29 3pm
  • Four companies, that are part of the 2nd Batch of this accelerator will do demo, followed a showcase of 20 cybersecurity and machine learning companies)
  • For more details, see

Tuesday, April 18, 2017 - 11:26

ML@GT Seminar by Manos Antonakakis (ECE) on “Using DNS & Machine Learning to Reason About Internet Abuse” at 12n on March 29, 2017, in Nano 1117/8

Speaker: Manos Antonakakis, School of Electrical and Computer Engineering (ECE), GA Tech

Location: Marcus Nanotechnology 1117-1118 (Map)

Date/Time: March 29, 2017, 12:00n – 1:00pm (Lunch at 11:30am)Title: Using DNS & Machine Learning to Reason About Internet Abuse

Title: Using DNS & Machine Learning to Reason About Internet Abuse

Abstract: The Domain Name System (DNS) is a critical component of the Internet. The critical nature of DNS often makes it the target of direct cyber-attacks and other forms of abuse. Cyber-criminals rely heavily upon the reliability and scalability of the DNS protocol to serve as an agile platform for their illicit network operations. For example, modern malware and Internet fraud techniques rely upon the DNS to locate their remote command-and-control (C&C) servers through which new commands from the attacker are issued, serve as exfiltration points for the information stolen from the victim’s computer and to manage subsequent updates to their malicious toolset.

In this talk, I will discuss how we can reason about Internet abuse using DNS and various machine learning methods. After providing an overview around DNS, botnets and their illicit activities, I will discuss how spectral methods can help us model one of the most agile threats on the Internet; the botnets that employ Domain Name Generation Algorithms (DGAs). Then, we will discuss ways that tensors can help us track virtual illicit actors across the Internet. Finally, I will conclude by discussing some open research problems in computer security where machine learning methods should be the key ingredient for any efficient and effective solution.

mantonakakisBio: Manos Antonakakis, Ph.D., is an Assistant Professor in the School of Electrical and Computer Engineering (ECE), and adjunct faculty in the College of Computing (CoC), at the Georgia Institute of Technology. He is responsible for the Astrolavos Lab, where students from both CoC and ECE conduct research in the areas of Network Security, Intrusion Detection, and Data Mining. In May 2012, he received his Ph.D. in Computer Science from the Georgia Institute of Technology. Before joining the ECE faculty, Professor Antonakakis held the Chief Scientist role at Damballa, where he was responsible for advanced research projects, university collaborations, and technology transfer efforts. He currently serves as the co-chair of the Academic Committee for the Messaging, Malware, and Mobile Anti-Abuse Working Group (M3AAWG). Dr. Antonakakis is the author of several U.S. patents and academic publications. He served as an external reviewer or a program committee member for leading information security conferences. He has successfully raised funding from multiple government agencies and organizations in the private sector. He is a member of the Institute for Information Security & Privacy (IISP) at Georgia Tech and contributed to its predecessor, the Georgia Tech Information Security Center (GTISC).

Tuesday, April 18, 2017 - 11:26

Ghassan Alregib’s Team (ECE) is hosting and running the IEEE VIP Cup in 2017 on the topic of “Traffic Sign Detection under Challenging Conditions”

Professor Ghassan Alregib’s team was been selected to be the team hosting and running the IEEE VIP Cup in 2017 on “Traffic Sign Detection under Challenging Conditions.” The topic is at the intersection of autonomous vehicles and machine learning. Not only it is attractive to the industry and the research communities but also the undergraduate students; the main goal is to get the UG students engaged with cutting-edge technologies. One of the side products is an open source dataset that has labeled ground truth data to test the vision algorithms under practical and real weather conditions, which have been missing in the publicly existing algorithms built for autonomous vehicles.vip_webpage_updated

Tuesday, April 18, 2017 - 11:26