Marilyn Walker

Last updated
Marilyn A. Walker
NationalityAmerican
CitizenshipUnited States of America
Alma mater University of Pennsylvania (Ph.D.)
Known for
Scientific career
Fields Computer science
Institutions University of California, Santa Cruz
University of Sheffield
AT&T Labs Research
Bell Labs
Thesis Informational Redundancy and Resource Bounds in Dialogue  (1993)
Doctoral advisor Aravind K. Joshi
Other academic advisorsKaren Sparck Jones, Ellen F. Prince
Website users.soe.ucsc.edu/~maw

Marilyn A. Walker is an American computer scientist. [1] She is professor of computer science and head of the Natural Language and Dialogue Systems Lab at the University of California, Santa Cruz (UCSC). Her research includes work on computational models of dialogue interaction and conversational agents, analysis of affect, sarcasm and other social phenomena in social media dialogue, acquiring causal knowledge from text, conversational summarization, interactive story and narrative generation, and statistical methods for training the dialogue manager and the language generation engine for dialogue systems.

Contents

Biography

Walker received an M.S. in Computer Science from Stanford University in 1987, and a Ph.D. in Computer and Information Science and an M.A in linguistics from the University of Pennsylvania in 1993. [2] Walker was awarded a Royal Society Wolfson Research Fellowship at the University of Sheffield from 2003 to 2009. She was inducted as a Fellow of the Association for Computational Linguistics (ACL) [3] in December 2016 for "fundamental contributions to statistical methods for dialog optimization, to centering theory, and to expressive generation for dialog". She served as the general chair of the 2018 North American Association for Computational Linguistics (NAACL-2018) conference.

Walker pioneered the use of statistical methods for dialog optimization at AT&T Bell Labs Research where she conducted some of the first experiments on reinforcement learning for optimizing dialogue systems. [4] Her research on Centering Theory is taught in standard textbooks on NLP. She also pioneered the use of statistical NLP methods for Natural Language Generation with the development of the first statistical sentence planner for dialogue systems in 2001. [5] She is well known for her work with François Mairesse on recognizing Big Five personality from text as well as using statistical methods for stylistic Natural Language Generation to express a particular Big Five personality type. [6] [7] An extension of this work learns how to manifest the linguistic style of a particular character in a film. [8]

She has published over 300 papers and is the holder of 10 U.S. patents. Her work on the evaluation of dialogue systems conducted at AT&T Bell Labs Research (PARADISE: A framework for evaluating spoken dialogue agents [9] ) is a classic, has been cited more than 1100 times. At UCSC, her lab focuses on computational modeling of dialogue and user-generated content in social media such as weblogs, including spoken dialogue systems and interactive stories. She led the Athena team, which was selected as a contender in the Alexa Prize SocialBot Challenge for 5 challenges between 2018 and 2023. [10]

References

  1. "dblp: Marilyn A. Walker". dblp.uni-trier.de. Retrieved 2019-12-31.
  2. "University of California Santa Cruz – Marilyn A. Walker". users.soe.ucsc.edu. Retrieved 2019-12-31.
  3. "ACL Fellows |".
  4. Walker, M. A. (2000-06-01). "An Application of Reinforcement Learning to Dialogue Strategy Selection in a Spoken Dialogue System for Email". Journal of Artificial Intelligence Research. 12: 387–416. doi:10.1613/jair.713. ISSN   1076-9757.
  5. Walker, Marilyn A.; Rambow, Owen; Rogati, Monica (2001). "SPoT: A Trainable Sentence Planner". Second Meeting of the North American Chapter of the Association for Computational Linguistics.
  6. Mairesse, F.; Walker, M. A.; Mehl, M. R.; Moore, R. K. (2007-11-28). "Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text". Journal of Artificial Intelligence Research. 30: 457–500. doi:10.1613/jair.2349. ISSN   1076-9757.
  7. "Controlling user perceptions of linguistic style: Trainable generation of personality traits". scholar.google.com. Retrieved 2025-09-27.
  8. "All the World's a Stage: Learning Character Models from Film". Seventh Artificial Intelligence and Interactive Digital Entertainment Conference. 2011.
  9. "PARADISE: A framework for evaluating spoken dialogue agents |". July 1997: 271–280. doi: 10.3115/976909.979652 . S2CID   3132651.{{cite journal}}: Cite journal requires |journal= (help)
  10. "Alexa Prize |".