CS Colloquium Series @ UCY
Department of Computer Science - University of Cyprus
The Department of Computer Science at the University of Cyprus holds research colloquiums and social hours approximately once weekly. All university students, faculty, and staff are invited to attend. Notifications about new and upcoming events are automatically disseminated to a variety of institutional lists.
If you don't receive these notifications, but want to get informed about upcoming colloquium announcements, you can do the following:
List
RSS
Directions
Colloquium Coordinator: Demetris Zeinalipour
Colloquium: Learning in a Partially Observable World, Dr. Loizos Michael (Open University of Cyprus, Cyprus), Monday, April 4, 2011, 11:00-12:00 EET.
The Department of Computer Science at the University of Cyprus cordially invites you to the Colloquium entitled:
Learning in a Partially Observable World
 |
Speaker: Dr. Loizos Michael |
Abstract:
Agents sensing their environment obtain information that is often incomplete in some shape or form. Examples abound: (1) certain tests may be too expensive to perform to complete a patient's medical record; (2) responders to a market survey may choose not to reveal certain information about themselves; (3) the author of a piece of text may choose not to explicitly state information that she believes can be inferred by the readers; (4) a packet may be routed through a private network and that route segment may not be tracked; (5) a dynamic system may transition through certain states too quickly to be monitored; (6) a user's preferences in support of a decision may be kept secret for privacy reasons. From a learning point of view, the challenge is to design algorithms that deal with incomplete information in a principled manner. We shall consider two broad settings: The static setting (examples 1-3) builds upon typical supervised learning scenarios, where, however, attributes are hidden in arbitrary ways. The dynamic setting (examples 4-6) deals with scenarios where an initial and a final state of a process are observed, while the intermediate states remain hidden. We shall discuss conditions under which algorithms are known to exist in these settings and can be shown to be efficient, be accompanied by predictive guarantees, and make limited assumptions on how information is hidden.
Short Bio:
Loizos Michael is a Lecturer in Information Systems at Open University of Cyprus (since 2009). Before joining OUC he held a Visiting Lecturer position at University of Cyprus (2008-2009). He was educated at University of Cyprus, where he received a B.Sc. in Computer Science with a minor degree in Mathematics (2002). He continued his education at Harvard University, where he received an M.Sc. and a Ph.D. in Computer Science (2003 and 2008, respectively). His research focuses on the formal and principled understanding of cognitive processes such as learning and reasoning, and how those are employed by humans and other biological organisms in their everyday lives. Specific areas of interest include: commonsense reasoning, temporal and default reasoning, computational learning theory, computational evolution theory, text and narrative understanding, nature-inspired computation, distributed computation, and game theory.
Sponsor: The CS Colloquium Series is supported by a generous donation from  |