The Physics of Information Technology
Abstract
The familiar devices that we use to collect,
transmit, and interact with electronic
information operate surprisingly close to very
many fundamental physical limits. Understanding
how such devices work, and how they can (and
cannot) be improved, requires deep insight into
the character of physical law as well as
engineering practices. The Physics of
Information Technology provides this needed
connection by introducing underlying governing
equations and then deriving operational device
principles.
About The Speaker
Prof Neil Gershenfeld leads the Physics and
Media Group at the MIT Media Lab and directs
the Things that Think research consortium. He
is the author of many books and has B.A.in
Physics from Swarthmore College and has a Ph.D
from Cornell University. He was a Junior Fellow
of the Harvard Society of Fellows, and a member
of the research staff at Bell Labs. He is
actively associated with Media Lab Asia
program.
|
Speech and Language Technologies
This talk will consist of two parts. In the
first part, an overview of research in the
Cognitive Machines Group at the MIT Media
Laboratory will be presented. The group is
developing machines that learn to converse
about what they see and do. In order to bridge
the continuous, noisy world of perception to
the symbolic world of natural language, the
group addresses the question of semantic
grounding: How are the meanings of words
connected to the physical world? Prof. Roy will
argue that current approaches to representing
the meaning of natural language are limited
since they fail to address the grounding
problem. The group's approach is to develop
models of grounding by building language
learning systems that acquire visually-guided
models of words and grammars. Prof. Roy will
present their experiments in visually-grounded
language acquisition by robots. He will also
discuss applications in automated verbal
descriptions of web sites, video content
retrieval by natural language query,
conversational robotics, and a variety of other
multimodal human-machine interfaces.
The second part of the talk is open-ended in
nature. Prof. Roy believes India is in a
position to push the envelope on speech and
language technology. Its highly multilingual
nature and high rates of illiteracy highlight
the importance of speech and multimodal
technologies. There is immense motivation to
solve technical problems associated with these
areas of language technology. This situation
raises numerous questions. How can the full
potential of the Indian technical and
linguistic communities be harnessed to solve
standing problems in speech and language
processing? What kinds of national research and
development infrastructure are necessary to
accelerate progress? Are there unique aspects
of the Indian context that can be leveraged to
create new classes of technologies? What roles
should language technologies play in education
in India? Please come join in on the
discussion!
About the speaker
Deb Roy is a Professor of Media Arts and
Sciences at the MIT Media Laboratory where he
founded and directs the Cognitive Machines
Group. Roy holds a Bachelor's Degree in
Computer Engineering from the University of
Waterloo, Canada and a Master of Science and
Ph.D. from MIT. Roy's current research focuses
on methods for building machines that learn to
communicate in human-like ways.
|
