Affiliated Faculty
 
 
     
Dr. Atkeson's research focuses on the application of machine learning to robotics and intelligent environments. he is interested in getting robots (particularly humanoid robots) to learn from their errors. Another goal is to build environments that learn to understand what people are doing, and learn how to help them more effectively.
 
     
 
Dr. Bagnell is interested in "closing the loop" on complex systems; that is, designing algorithms that allow systems to observe their own operation and improve performance. He is currently focused on applications of learning and decision making applied to mobile robotics and developing rich, structured models that are appropriate for both making and learning decisions.
 
 
 
Dr. Blum's main research interests are in Machine Learning theory and on-line algorithms. His work involves designing algorithms with provable performance guarantees, as well as developing new models for analyzing emerging problems such as learning from labeled and unlabeled data.
 
 
  Dr. Carbonell's research interests span several areas of artificial intelligence and data mining, including: inductive machine learning methods, natural language processing, machine translation, information retrieval, fact extraction from the web and from free text, and automated summarization (where he invented MMR search-diversity technology).
 
 
  Dr. Dey is interested in applications of machine learning to ubiquitous computing, and, in particular, to systems that can sense the environment around them and help users make decisions about appropriate actions to take. This includes the design of smart environments that can determine what actions are taking place and creating models and detecting trends in different aspects of human behavior. Current domains of interest are healthcare, driving and smart homes.
   
   
 

Dr. Dubrawski's research interests are in autonomous systems that work, are useful and make economic sense, and in finding ways to effectively build and deploy them.
   
   
 
Dr. Fahlman has worked in many areas of AI: planning, knowledge representation, image processing, natural language, document classification, artificial neural networks, and the use of massively parallel machines in AI. Currently he is working on a practical, easy to use system for symbolic knowledge representation and is exploring ways to combine symbolic and statistical methods.
 
 
 
Dr. Glymour's current research applies previous work on causal Bayes nets and formal learning theory to a variety of topics. With collaborators at NASA Ames he works on automated identification of mineral composition from spectra. With the Computational Systems Biology Group, he works on the possibilities and limitations of machine learning procedures for inferring gene regulation from measurements of messenger RNA concentrations. In collaboration with several psychologists he also works on mathematical aspects of the psychology of causal reasoning.His current work also concerns predictions of biosphere events (e.g., forest fires) from satellite measurements of spectra. With Joseph Ramsey he is preparing Bayes net software for deployment on a mission to Mars scheduled for 2009.
 
     
  Dr. Seth Copen Goldstein's research focuses on computing systems and nanotechnology. He believes that the fundamental challenge for computer science in the twenty-first century is how to effectively harness systems which contain billions of potentially faulty components. One of the projects he works on that addresses this issue is the Claytronics project, which is exploring the hardware and software necessary to realize programmable matter.
   
 
 
Charles Kemp works on statistical models of human learning and cognitive development. His interests include concept learning, common-sense reasoning, and other problems that are readily solved by people but difficult for machines to handle.
 
 
 

Dr. Koedinger is interested in the use and advancement of machine learning as a tool for modeling human learning, for creating simulated students, for accelerating development of intelligent tutoring systems, and data mining of student interactions in e-learning environments.
 
 
 
Mike Lewicki's research is centered around the problem of representing and learning structure in natural auditory and visual environments. His research employs a broad range of disciplines including statistical learning theory, machine vision and audition, signal processing, and computational neuroscience.
 
 
 
Dr. Liu's research focus is on learning semantically discriminative image features from large image datasets, especially biomedical images and their collateral information. Her computational tools are drawn from statistical learning theory, group theory, computer vision and pattern recognition. The goal of her research is to seek the intrinsic dimensionality and separability in a large amount of labeled and unlabeled images.
 
 
 

Dr. Mostow is founder and Director of Project LISTEN (http://www.cs.cmu.edu/~listen), which is using computers to listen to children read aloud.  Project LISTEN’s Reading Tutor serves as a delivery vehicle for one-on-one instruction adapted to the individual student, as a richly instrumented platform to collect large amounts of fine-grained, longitudinal data, and as a research tool to carry out educational experiments invisibly embedded in the Reading Tutor with many thousands of randomized controlled trials.  Reading Tutor databases from successive school years offer opportunities for innovative work in educational data mining.
 
 
 
Dr. Neill's research interests are in statistical machine learning, data mining, and pattern detection. He is particularly interested in developing new statistical and computational methods for the early, automatic detection of emerging public health threats ranging from avian influenza to bioterrorism. He is also investigating a variety of other large-scale anomaly detection problems related to medicine, public health, and homeland security.
 
 
 
Dr. Sandholm's research interests are in active learning, stochastic optimization, electronic commerce; game theory; mechanism design; artificial intelligence; multiagent systems; auctions and exchanges; automated negotiation and contracting; voting; coalition formation; safe exchange; search, integer programming and combinatorial optimization; preference elicitation; normative models of bounded rationality; resource-bounded reasoning; privacy; multiagent reinforcement learning.
 
 
 
Dr. Schneider's research interests are in Machine Learning, reinforcement learning, optimization, and decision making. He has applied his methods to business applications ranging from process control, to production scheduling and inventory management, to long range strategic planning.
 
 
 
Dr. Smith's research focuses on the application of statistical modeling to problems in natural language processing. His interests include predicting the linguistic structure in multilingual text (morphology, syntax, and semantics), learning linguistic structure from unannotated corpora, and building robust models for applications
like machine translation and question answering.
 
 
  Luis von Ahn's research interests include: novel techniques for
utilizing the computational abilities of humans, such as games in
which people collectively solve large-scale problems that computers
cannot yet solve (e.g., http://www.espgame.org, http://www.peekaboom.org); human-computer interaction, artificial intelligence, and the difference in computational abilities between humans and computers; theoretical cryptography and security, and computer science theory in general.