|
Making
Sense of Meaning - New Computational Perspectives
Professor
Hari Sundaram Dept. of Computer Science
Arizona State University
Friday, May 16, 11am SITE 5084, 800 King
Edward Ave
Sponsored
by
DISCOVER Lab
| Abstract
The problem of the "semantic gap," - how to resolve sensory data to
meaning, is a familiar and significant challenge within the media
computing and AI communities. Today, a majority of the work in
resolving the semantic gap lies in the learning paradigm, where our
knowledge of the domain is encoded in terms of identification of
effective features and appropriate learning algorithms. This talk
will offer two long term research perspectives to address the
semantic gap, where computational learning is integral, but which
expand the problem space.
The first perspective is in integrating multimedia with the physical
world. The important consequence of this idea is that we move away
from offline analysis of media semantics. In this idea, semantics are
jointly constructed by the media system and the user, within a
real-time, mediated, feedback control loop. I shall present our work
in stroke rehabilitation as a specific example of this class of
research. The goal of this research is to transfer the specific
movement semantics, to enable stroke patients to perform challenging
functional tasks ("reach and grasp a cup"). The second perspective is
recognizing that meaning is an emergent, evolving artifact of
collaborative human activity. In this idea, identification of human
networks that produce meaning is a critical first step. I will
present community discovery in large scale blog networks as a
concrete example of this research.
Both long term research frameworks provide us with a fresh set of
computational problems that are synergistic with the learning
paradigm. I shall briefly present other examples - optimal resource
constrained real-time decision making, information flow in online
networks, and collaborative annotation in data sparse media
collections.
|
| Biography of
Hari Sundaram
Hari Sundaram is currently an assistant professor of media arts and
computing, with the Arts Media and Engineering program, and Computer
Science at Arizona State University. He received his Ph.D. from the
Department of Electrical Engineering at Columbia University in 2002.
He received his MS degree in Electrical Engineering from SUNY Stony
Brook 1995 and a B.Tech in Electrical Engineering from Indian
Institute of Technology, Delhi in 1993.
His intellectual commitment is towards understanding how meaning
emerges through our engagement with the physical online worlds. His
research has focused on two complementary (but coupled) directions -
(a) designing intelligent media environments that exist as part of
our physical world (e.g. mediated environments that assist stroke
patients recover) (b) developing new algorithms and systems to
understand the media artifacts resulting from human activity (e.g.
emails, photos / video). Specific projects include - context models
for interpreting human action, understanding communication patterns
in media sharing social networks, collaborative annotation, as well
discovering emergent groups in online social networks.
His research has won several awards - the best student paper award at
JCDL 2007, the best ACM Multimedia demo award in 2006. The best
student paper award at ACM Multimedia 2002, the 2002 Eliahu I. Jury
Award for best Ph.D. dissertation. He has also received a best paper
award on video retrieval from IEEE Trans. On Circuits and Systems for
Video Technology, 2000.
He is an active participant in the Multimedia community - he is an
associate editor for ACM Transactions on Multimedia Computing,
Communications and Applications (TOMCCAP), as well as the IEEE Signal
Processing magazine. He has co-organized workshops at acm multimedia
on experiential telepresence (ETP 2003, ETP 2004), archival of
personal experiences (CARPE 2004, CARPE 2005) and a conference of
image and video retrieval (CIVR 2006).
This website is maintained by Yongyi Mao, and was last
updated on April 30, 2008
|