Raymond Lau
 
HOME | WORK | LEARN | EAT | LIKE   
On this page, you will a details of Ray's academic background, grouped into sections ordered in reverse, chronologically. You will also find a complete listing of his academic publications.

Post-Graduate

Subsequent to completing his doctorate, Raymond spent just short of two years (April 1998 to December 31, 1999) as a Research Scientist in the Spoken Language Systems Group at the MIT Laboratory for Computer Science (now part of the MIT Computer Science and Artificial Intelligence Laboratory). Lau was one of the two principal architects (together with Stephanie Seneff) responsible for the GALAXY-II framework for conversational systems development, building upon his earlier work on WebGALAXY (See Undergraduate and Graduate below). GALAXY-II has been designated as the initial reference architecture for the DARPA Communicator program, a major research and development effort in the area of dialogue-based intelligent conversational interfaces. There is an ICSLP 98 paper describing GALAXY-II and the code from that project has since been made available via open source here. Lau also worked on the Mercury system for retrieving information about flight schedules and pricing via telephone.

Undergraduate and Graduate

Ray received his Bachelor of Science degree in Computer Science and Engineering from the Massachusetts Institute of Technology in 1993. He received the Masters of Science degree in Electrical Engineering and Computer Science also from MIT in 1994. His masters research was in the area of adaptive statistical language models of English and the work was performed at IBM's Thomas J. Watson Research Center under the directions of Dr. Salim Roukos at IBM and Professor Victor W. Zue at MIT.

In 1998, he received the Ph.D. in computer science from MIT as well. Lau performed his doctoral work in the Spoken Language Systems Group at the MIT Laboratory for Computer Science (now part of the MIT Computer Science and Artificial Intelligence Laboratory). Lau's doctoral dissertation, completed in March 1998, was in the area of subword linguistic modelling for speech recognition, that is, using word substructures such as syllables for better recognition, at SLS under the direction of Dr. Stephanie Seneff. While pursuing his doctorate, Ray also led the WebGALAXY project. Lau's research interests included: voice recognition, conversational language systems, language modelling, and various Internet related technologies. Lau was a National Science Foundation fellowship recipient and is a member of Eta Kappa Nu.

Primary and Secondary

Raymond attended Public School 69 in Jackson Heights (Queens, NYC) for grades K through 5 and Joseph Pulitzer Intermediate School 145, also in Jackson Heights, for grades 6 through 8. He was a trumpet player in the I.S. 145 band, was a member of the math team, and graduated as valedictorian in 1985. His validictory speech focused on the computing revolution as the major development for the next decade.

In 1989, Ray received his high school diploma from Stuyvesant High School with honors. He was a semi-finalist in the prestigious Westinghouse Science Talent Search (now the Intel Science Talent Search) for the 1988-1989 academic year with a submission in the area of optical character recognition (OCR).

Academic Publications

There are three groupings of publications. The first is related to ANGIE and Lau's Ph.D. dissertation work. The second is related to GALAXY and other work done for the Spoken Language Systems Group. The third is related to trigger-based adaptive language modelling work for Ray's S.M. done at IBM TJ Watson Research Center.

ANGIE and Ph.D. Work Publications

A Unified System for Sublexical and Linguistic Modelling Suporting Flexible Vocabulary Speech Understanding Authors: Raymond Lau and Stephanie Seneff
Appears in: Proceedings of ICSLP '98, Sydney, Australia, pp. 2443-2446, November, 1998
Subword Lexical Modelling for Speech Recognition Author: Raymond Lau
Ph.D. thesis, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, May 1998
Providing Sublexical Constraints for Word Spotting within the ANGIE Framework Authors: Raymond Lau and Stephanie Seneff
Appears in: Proceedings of Eurospeech '97, Rhodes, Greece, pp. 263-266, September, 1997

ANGIE: A New Framework for Speech Analysis Based on Morpho-Phonological Modelling Authors: Stephanie Seneff, Raymond Lau and Helen Meng
Appears in: Proceedings of ICSLP '96, Philadelphia, PA, pp. 110-113, October, 1996

Excerpt of 1996 SLS Annual Research Summary

A more digestible excerpt describing this work is also available here.

GALAXY and SLS Work Publications

Organization, Communication, and Control in the GALAXY-II Conversational System Authors: Stephanie Seneff, Raymond Lau, and Joseph Polifroni
Appears in: Proceedings of Eurospeech '99, Budapest, Hungary, pp. ??-??, September, 1999
Galaxy-II: A Reference Architecture for Conversational System Development Authors: Stephanie Seneff, Ed Hurley, Raymond Lau, Christine Pao, Philipp Schmid and Victor Zue
Appears in: Proceedings of ICSLP '98, Sydney, Australia, pp. 931-934, November, 1998
WebGALAXY: Integrating Spoken Language and Hypertext Navigation Authors: Raymond Lau, Giovanni Flammia, Christine Pao and Victor Zue
Appears in: Proceedings of Eurospeech '97, Rhodes, Greece, pp. 883-886, September, 1997

WebGALAXY: Beyond Point and Click - A Conversational Interface to a Browser Authors: Raymond Lau, Giovanni Flammia, Christine Pao and Victor Zue
Appears in: M. R. Genesereth and A. Patterson (eds)., Proceedings for the Sixth International World Wide Web Conference, Santa Clara, CA, pp. 119-128, April, 1997

S.M. Work at IBM TJWRC Publications

Adaptive Statistical Language Modelling Author: Raymond Lau
S.M. Thesis, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, May 1994.

Trigger-based Language Models: A Maximum Entropy Apporach Authors: Raymond Lau, Roni Rosenfeld and Salim Roukos
Appears in: Proceedings ICASSP '93, Minneapolis, MN, pp. II-45 - II-48, April, 1993.
(Note: Sorry, only images are available.)

Adaptive Language Modelling Using the Maximum Entropy Approach Authors: Raymond Lau, Roni Rosenfeld and Salim Roukos
Appears in: Proceedings ARPA Human Language Technologies Workshop, Princeton, NJ, pp. 81-86, March, 1993.
(Note: Sorry, only images are available.)

Building scalable N-gram language models using maximum likelihood maximum entropy N-gram models U.S. Patent: 5,467,425
Inventors: Lau, Raymond; Rosenfeld, Roni; Roukos, Salim
Assignee: International Business Machines Corporation (Armonk, NY)
Filed: Feb. 26, 1993
Issued: Nov. 14, 1995
21 Claims, 16 Drawing Figures

Certificate of correction issued on Jun. 17, 1997, as U.S. Patent 5,640,487

Abstract:

The present invention is an n-gram language modeler which significantly reduces the memory storage requirement and convergence time for language modelling systems and methods. The present invention aligns each n-gram with one of "n" number of non-intersecting classes. A count is determined for each n-gram representing the number of times each n-gram occurred in the training data. The n-grams are separated into classes and complement counts are determined. Using these counts and complement counts factors are determined, one factor for each class, using an iterative scaling algorithm. The language model probability, i.e., the probability that a word occurs given the occurrence of the previous two words, is determined using these factors.

Full list of claims and images from the USPTO

Excerpt of 1995 SLS Annual Research Summary

A more digestible excerpt describing this work is also available here.

 > Last Updated: December, 2004> Contact: mail@raylau.com