Thomas G. Dietterich

E-mail: tgd@cs.orst.edu Phone: +1-541-737-5559 Office: KEC 2067
PGP Public Key Recent changes to this page. (Last updated May 5, 2008.)

Page Contents: Research Prospective Students Publications CV Software Students and Staff Course Materials Conferences

Current Research

"If you invent a breakthrough in artificial intelligence, so machines can learn," Mr. Gates responded, "that is worth 10 Microsofts." (Quoted in NY Times, Monday March 3, 2004)

The focus of my research is machine learning: How can we make computer systems that adapt and learn from their experience? How can we combine machine learning with other advances in AI to build Integrated Intelligent Systems? How can we combine human knowledge with massive data sets to expand scientific knowledge and build more useful computer applications? My laboratory combines research on machine learning and AI fundamentals with applications to problems in science and engineering.

• Scientific Projects
  • Ecosystem Informatics: Oregon State University is a leader in combining computer science and the ecological sciences to build the new discipline of Ecosystem Informatics. Ecosystem Informatics studies methods for collecting, analyzing, and visualizing data on the structure and function of ecosystems. It is an instance of an important new direction in science: Data Exploration Science (see Jim Gray's 2003 KDD talk).

    My group is involved in many Ecosystem Informatics activities:

    • Insect Identification. Our current understanding of complex ecosystems is limited by a lack of data. One particularly useful kind of data is population counts of "bugs" (small athropods that live in soils, lakes, streams, and the ocean). We seek to develop devices for capturing, imaging, and sorting bugs combined with general image processing/machine learning/pattern recognition tools for counting and classifying them. We hope to transform the ability of scientists to measure the health of forests, streams, and estuaries. More generally, we are interested in developing a wide range of novel instruments for expanding the quality, quantity, and spatio-temporal resolution of ecologically-relevant data. We recently received a grant of $800,000 from the NSF to support our work in this area. See OSU News Service Story. Project Pages. Image Databases.

    • Automated Data Cleaning for Sensor Networks. As large scale sensor networks are deployed to collect ecological data, methods are needed for automated quality assurance ("data cleaning") for this data to detect and remove errors due to sensor failure. Graduate student Ethan Dereszynski has developed Dynamic Bayesian Network (DBN) models of sensor network data from the Andrews Experimental Forest. These models are being applied to automatically detect and flag anomalies in temperature data.

    • Machine Learning for Species Distribution. One of the central goals of ecology is to understand and predict the distribution of species (including the bugs that we are studying in the Insect Identification project). Given a data set that records observations of the presence (or absence) of multiple species at multiple locations, we wish to develop models that can predict their presence/absence elsewhere. Graduate student Twinkle Lettkeman is developing novel learning algorithms for this task.

    • IGERT (Integrative Graduate Education and Research Traineeships). We are recipients of an NSF IGERT grant that funds Traineeships in Ecosystem Informatics. The goal is to train PhD students for interdisciplinary research that combines computer science, mathematics, and ecosystem science to prepare graduate students for interdisciplinary research in these areas. Ph.D. students in this program receive a stipend of up to $30,000 per year for 2 years (plus a spending allowance of $10,000 for computers, travel to conferences, paid internship, etc.) Support continues after the 2-year period on individual research grants. Information for Prospective IGERT Students.

    • Summer Institute in Ecosystem Informatics. Junior and Senior undergraduates and first-year graduate students can spend 10 weeks studying and researching at the H. J. Andrews Experimental Forest under funding from this NSF summer program. As with the IGERT, the goal is to introduce students to the research challenges at the intersection of mathematics, computer science, and the ecosystem sciences.

  • Intelligent Desktop Assistants. We are involved in two large efforts to develop intelligent assistants for the computer desktop.

    • TaskTracer. When you come into work in the morning, you don't want to say to your computer "I want to run Word", but rather, "I want to work on my CS534 homework". In other words, you would like a user interface that was organized around your projects and activities rather than around application programs, files, folders, etc. You would also like all of your information in one place rather than scattered across the local file system, network file systems, web sites, email folders, calendar, contacts, etc. TaskTracer extends the Windows UI to provide exactly this functionality. This research is supported by DARPA through the CALO project and gifts from Intel. OSU News Service story. Project Web Site. TaskTracer is also in the process of being commercialized at Smart Desktop.

    • CALO. The CALO project seeks to develop an AI personal assistant that can help you find relevant documents, prepare for meetings, keep track of what is going on during meetings, and autonomously execute tasks such as arranging travel, scheduling meetings, executing administrative workflows (e.g., purchasing and staffing), and so on. We have four roles in the CALO project. First, we are managing the CALO Project called Organize and Prepare Information. Second, we are managing two activities within CALO: Integrating Learning and Reasoning (ILR) and Activity Recognition and Proactive Assistance (ARPA). Third, we coordinate work on Transfer Learning. Finally, we are incorporating parts of TaskTracer in the CALO system.

  • AI for Computer Games. The AI components of most computer games are hand-authored rule-based systems. We are studying methods for developing game AI agents via machine learning. One approach is to apply reinforcement learning to automatically learn an opponent for games. Another approach is to teach the game AI by demonstrating how to play the game. A third approach is to transfer knowledge learned on one game to rapidly construct an AI agent for a new game. We use the RTS game Wargus as our experimental platform. Wargus is based on the Strategus game engine. This work is funded by DARPA under the Transfer Learning program and by ARO under a MURI grant.

• Fundamental Machine Learning Research
  • Sequential and Spatial Supervised Learning (SSSL). Many emerging applications of machine learning involve sequential or spatial data including the scientific applications listed above as well as problems of transaction monitoring, counter-terrorism, and fraud detection. Most present-day applications of machine learning to spatio-temporal require extensive ad hoc tool-building. Can we build a new generation of generic machine learning tools that can be applied "off the shelf" to solve these sequential and spatial supervised learning problems?

  • Transfer Learning. In collaboration with researchers at Berkeley, Stanford, and MIT, we are studying new learning methods that can transfer learned knowledge from one setting or context to another. This requires learning to occur at a more abstract and more relational level than in standard supervised learning algorithms. Our primary application area is learning in Wargus (see above).

  • Integrated Learning. We are studying ways to combine reasoning and multiple knowledge sources to learn hierarchical task knowledge from a single demonstration of a task coupled with rich interaction. Our primary application area for this project is air traffic flight planning.

• Reviews, tutorials, and books. I have written several review articles and tutorials on machine learning.

There is a weekly AI Colloquium where we present and discuss research we are doing here at Oregon State and where visitors also make presentations from time to time. Students interested in joining my group are strongly encouraged to attend these meetings.

Information for Prospective Students

If you are seeking a research career in machine learning, data mining, artificial intelligence and related areas, and you have a strong background in mathematics and programming, please read my Information for Prospective Students page. To see what courses I expect my Ph.D. students to take, please see Recommended Courses for Ph.D. Students in Machine Learning.

Publications, Curriculum Vita, and Software

• Publications
• Talks
• Curriculum Vita
• Downloadable Collection of Error Correcting Codes for use with the error-correcting output coding technique.
• Release 1.0 of MAXQ Hierarchical Reinforcement Learning code.
• RSW (Recurrent Sliding Window) Package for WEKA for sequential supervised learning.
• Implementation of the PCBR region detector developed as part of the BugID project.

Journals and Book Series

• I am a member of the Advisory Board for the Journal of Machine Learning Research, which is an electronic (and hardcopy) journal covering all areas of machine learning.
• I serve on the Board of Advisors of Neural Computing Surveys.
• I edit the MIT Press book series on Adaptive Computation and Machine Learning.
• With Ron Brachman, I co-edit the Morgan-Claypool Synthesis Series on Artificial Intelligence and Machine Learning.
• I serve on the Board of Evaluation of Intelligent Systems, an online resource on methods for statistical evaluation of AI systems.
• I am past President of the International Machine Learning Society.

Entrepreneurial Activities

• I am a co-founder of Strands (formerly MyStrands; formerly MusicStrands), a recommendation company.
• I am a co-founder and Chief Scientist of Smart Desktop, an EMC company. Smart Desktop is commercializing the TaskTracer system.

Current Students and Staff

• Xinlong Bao, Graduate Student.
• Ethan Dereszynski, Graduate Student.
• Guohua Hao, Graduate Student.
• Jed Irvine, Software Developer.
• Victoria Keiser, Graduate Student.
• Twinkle Lettkeman, Graduate Student.
• Jianqiang Shen, Graduate Student.
• Michael Slater, Project Manager.
• Wei Zhang, Graduate Student.

Former Students and Staff

• Hussein Almuallim, Associate Professor, Computer Science, King Fahd University of Petroleum and Minerals; also with Veritas DGC (Oil Reservoir Modeling Company).
• Eric Altendorf, Google.
• Adam Ashenfelter, Strands, Inc., Corvallis, Oregon.
• Ghulum Bakiri, Assistant Professor, Computer Science, Bahrain University
• Brian Breck.
• Waranun Bunjongsat.
• Giuseppe Cerbone, CEO, ANSA (Agenzia Nazionale Stampa Associata), Italian news agency.
• Martha Chamberlin.
• Hei Chan.
• Richard Charon.
• Eric Chown, Associate Professor, Bowdoin College.
• Dan Corpron
• Diane Damon, Portland, OR.
• Phuoc Do, Smart Desktop.
• Nicholas Flann Associate Professor, Utah State University
• Greg Foltz.
• Dan Forrest
• Tony Fountain, Staff Scientist, San Diego Supercomputer Center.
• Ashit Gandhi, Founder and Vice-President, Prism Gem, LLC - The Art of Diamond Coloring.
• Colin Gerety, Fort Collins, CO.
• Brandon Harvey, Symantec, Eugene, OR.
• Hermann Hild, President, SMI Cognitive Software GmbH .
• Saket Joshi, Graduate Student, Tufts University.
• Varad Joshi, Interconnectix, Portland, OR.
• Caroline Koff, Hewlett-Packard Corporation, Fort Collins, CO.
• Michael Kelm, Ph.D. student, Interdisciplinary Center for Scientific Computing of the Ruprecht-Karls-University of Heidelberg
• Eun Bae Kong, Associate Professor and Department Head, Computer Science, Chungnam National University, South Korea
• Bill Langford, Post-doc at RMIT, Melbourne, Australia.
• Dragos Margineantu, The Boeing Company.
• Prafulla Mishra
• Avis Ng, Newbridge Networks, Inc. (now Alcatel)
• Soumya Ray, Assistant Professor, Case-Western Reserve University.
• Angelo Restificar, ZetaInteractive.
• Ritchey Ruff, Certification Consultant and Lead Tester at Interactive Home Systems.
• Rongkun Shen.
• Shriprakash Sinha
• Simone Stumpf. Whitehorse.
• Dan Vega
• Mark Vulfson. Microsoft Corporation.
• Xin Wang, Microsoft Corporation.
• Dietrich Wettschereck. Recommind.com.
• Pengcheng Wu.
• Michael Wynkoop, Qualcomm.
• Wei Zhang, The Boeing Company.
• Valentina Zubek, Aureon Biosciences.

Previous Courses and Courseware

• CS519/GEO599: Principles of Ecosystem Informatics, 2004-2005.
• CS 534, Spring 2005, Machine Learning.
• CS430, Fall 2003, Introduction to Artificial Intelligence
• CS539, Fall 2003, Seminar: Probabilistic Relational Models
• CS 533, Applied Artificial Intelligence for Engineeers.
• CS 539, Winter 2000, Selected Topics in Artificial Intelligence: Probabilistic Agents
• CS 430/530, Fall 1999, Artificial Intelligence Programming Techniques.
• CS 519, Fall 1996. Research Methods in Computer Science.
• CS 450/550, Winter 1996, Introduction to Computer Graphics.

Machine Learning Resources

• How to be a Graduate Student. A great web page with pointers to lots of good resources for graduate students.
• Standard Proofreading Symbols that I use when marking corrections on papers.
• Research Index at Penn State. An invaluable resource for finding online articles, citations, etc.
• Journal of Machine Learning Research (JMLR). (Free electronic and hardcopy journal.)
• Machine Learning. (Expensive hardcopy journal.)
• Journal of Artificial Intelligence Research (JAIR). (Free electronic and hardcopy journal.)
• The Machine Learning Database Repository at UC Irvine.
• The Machine Learning Programs Repository at UC Irvine.
• UC Irvine ML information.
• StatLib containing data, algorithms, and other information relevant to statistics.
• Reinforcement Learning Repository at UMass Amherst.
• Knowledge Discovery Mine containing information about knowledge discovery in databases.
• NIPS Online Proceedings.
• Neuroprose Archive containing recent papers relevant to neural networks.
• Machine Learning Information Services .
• CMU Reinforcement learning group.
• NeuroCOLT home page.
• List of Usenet FAQ's
• Bibliographies on Artificial Intelligence
• Monash University Minimum Message Length group
• Computational Intelligence Bibliography Collection.
• Numerical Recipes Homepage.

Conferences and Workshops

Upcoming Conferences

NIPS 2008, Vancouver and Whistler, BC, December 8-13, 2008.

Recent Conferences

International Conference on Machine Learning (ICML-2008). July, 2008. Helsinki, Finland.

AAAI-2008. July, 2008, Chicago, IL.

Neural Information Processing Systems (NIPS-21). December 3-8, 2007, Vancouver, British Columbia and Whistler, Canada.

My Family's Musical Activities

• The Stack. My son Noah writes songs and plays rhythm guitar in this rock-and-roll band. He also arranges lots of songs for Mens' Blue and White (Men's A Capella at the Claremont Colleges); watch them on YouTube.

• Jubilate: The Women's Choir of Corvallis. My wife Carol sings in this choir.

Tom Dietterich, tgd@cs.orst.edu