Eugene Charniak: 1. Personal
Eugene Charniak: 1. Personal
Eugene Charniak: 1. Personal
Curriculum Vitae
1. Personal
Professor
Department of Computer Science
Brown University
Providence, RI 02912
2. Home Address
106 Halsey Street
Providence, RI 02906
3. Education
4. Professional Appointments
1973-1975 Research Scientist, Institute of Semantic and Cognitive Studies, Lugano, Switzerland
1976-1977 Institute of Semantic and Cognitive Studies became part of the Computer Science
Department, University of Geneva
1977-1978 Visiting Assistant Professor, Department of Computer Science, Yale University, New
Haven, CT
1979-1981 Assistant Professor of Computer Science and Cognitive and Linguistic Science, Brown
University, Providence, RI
1981-1984 Associate Professor of Computer Science and Cognitive and Linguistic Science, Brown
University, Providence, RI
1984- Professor of Computer Science and Cognitive and Linguistic Science, Brown University,
Providence, RI
1
1984-1985 Visiting Professor of Computer Science, Yale University, New Haven CT
1998 Visiting Professor of Computer Science and Electrical Engineering, Johns Hopkins
University, Baltimore, MD
5. Publications
Books
Chapters in Books
Context and the Reference Problem,'' Natural Language Processing (ed. R. Rustin),
pp. 311-331, Algorithmics Press, New York (1972).
2
``Probabilistic Text Understanding'' (with Robert Goldman), Statistics and Computing
.Also in Artificial Intelligence Frontiers in Statistics:
AI and Statistics III (ed. D. J. Hand), Chapman and Hall (1992).
``A Statistical Syntactic Disambiguation Program and What it Learns'' (with Murat
Ersan), Symbolic, Connectionist, and Statistical Approaches to Learning for
Natural Language Processing (ed. S. Wermter, E. Riloff, and G. Scheler),
Springer, pp. 146-160 (1996).
``The Case-Slot Identity Theory,'' Cognitive Science 5 (3), pp. 285-292 (1981).
``Statistical Techniques for Natural Language Parsing,'' AI Magazine, 18 (4), pp. 33-43
(1997).
``New Figures of Merit for Best-first Probabilistic Chart Parsing'' (with Sharon
Caraballo), Computational Linguistics, 24, pp. 275-298 (1998).
3
Refereed Conference Articles
``With a Spoon in My Hand This Must be the Eating Frame,'' Proc. Second Workshop on
Theoretical Issues in Natural Language Processing (1978). Also published in
Journal of Computational Linguistics (1978).
``Word Sense and Case Slot Disambiguation'' (with Graeme Hirst), Proc. 1982 National
Conference on Artificial Intelligence (1982).
``The Bayesian Basis of Common Sense Medical Diagnosis,'' Proc. 1983 National
Conference on Artificial Intelligence (1983).
``Time and Tense in English'' (with Mary Harper), Proc. 1986 Conference of the
Association for Computational Linguistics, pp. 3-9 (1986).
``A Neat Theory of Marker Passing,'' Proc. 1986 National Conference on Artificial
Intelligence, pp. 584-589 (1986).
`A Logic for Semantic Interpretation'' (with Robert Goldman), Proc. 1988 Conference of
the Association for Computational Linguistics, pp. 87-94 (1988).
`A Probabilistic ATMS for Plan Recognition'' (with Robert Goldman), Workshop on Plan
Recognition, AAAI Press (1988).
4
``Plan Recognition in Stories and in Life'' (with Robert Goldman), Proc. Fifth Workshop
on Uncertainty in Artificial Intelligence, pp. 54-59 (1989).
``A New Algorithm for Finding MAP Assignments to Belief Networks'' (with
Solomon E. Shimony), Proc. Conference on Uncertainty in Artificial Intelligence,
(1990).
``A Probabilistic Model of Plan Recognition'' (with Robert Goldman), Proc. 1991
Conference of the American Association for Artificial Intelligence (1991).
``A New Admissible Heuristic for Minimal-Cost Proofs'' (with Saadia Husain),
Proc. 1991 Conference of the American Association for Artificial Intelligence (1992).
``Statistical Parsing with a Context-free Grammar and Word Statistics'' Proc. of the
Fourteenth National Conference on Artificial Intelligence AAAI Press/MIT Press (1997),
Best Paper Award.
5
``Edge-based Best-first Chart Parsing'' (with Sharon Goldwater and Mark Johnson),
Proc. Sixth Workshop on Very Large Corpora, pp. 127-133
(1998).
``A Statistical Approach to Anaphora Resolution'' (with Niyu Ge and John Hale),
Proc. Sixth Workshop on Very Large Corpora, pp.161-171 (1998).
``Assigning function tags to parsed text'' (with Don Blaheta) Proceedings of the 2000
Conference of the North American Chapter of the Assocation for Computational
Linguistics, ACL, New Brunswick NJ (2000)
"Immediate-head parsing for language models" Proceedings of the Association for 2001
(2001 Computational Linguistics) (Best paper award)
"Edit detection and parsing for transcribed speech" (with Mark Johnson), Proceedings of
the North American Association for Computational Linguistics 2001, pp. 118-126 (2001)
"Entropy Rate Constancy in Text" (with Dmitriy Genzel) Proceedings of the Association
for Computational Linguistics 2002
"Parsing and Disfluency Placement" (with Donald Engel and Mark Johnson)
Proceedings of the Conference on Empirical Methods in Natural Language Processing
Syntax-based Language Models for Statistical Machine Translation with Kevin Knight
and Kenji Yamada, Machine Translation Summit IX, 2003.
6
Variation of Entropy and Parse Trees of Sentences as a Function of the Sentence
Number with Dmitriy Genzel, 2003 Conference on Empirical Methods in Natural
Language Processing.
Technical Reports
``Micro-Planner Reference Manual'' (with G.J. Sussman and T. Winograd), A.I. Memo
203A, M.I.T. Artificial Intelligence Laboratory (1971).
``He Will Make You Take it Back, A Study in the Pragmatics of Language,'' Technical
Report 6, Institute of Semantic and Cognitive Studies, Castagnola, Switzerland (1974).
``A Brief on Case,'' Technical Report 22, Institute for Semantic and Cognitive Studies,
Castagnola, Switzerland (1975).
``On the Referential/Attributive Distinction,'' Technical Report 24, Institute for Semantic
and Cognitive Studies, Geneva, Switzerland (1976).
7
``The Frail/Nasl Reference Manual'' (with Michael K. Gavin and James Hendler),
Technical Report No. CS-83-06, Department of Computer Science, Brown University
(1983).
``A New Admissible Heuristic for Minimal-Cost Proofs'' (with Saadia Husain),
Technical Report No. CS-91-11, Department of Computer Science, Brown University
(1991).
``Parsing with Context-Free Grammars and Word Statistics,'' Technical Report No.
CS-95-28, Department of Computer Science, Brown University (1995).
Book Reviews
``How to Register Dissatisfaction with AI,'' \fIThe Behavioral and Brain Sciences 1 (1)
(1979).
Invited Lectures
8
1972 Stanford University
Rutgers University
1981 Hewlett-Packard
Bolt, Beranek and Newman
9
Ohio SIGART and Air Force Institute of Technology
10
1994 University of Toronto
Carnegie-Mellon University
Brown University Department of Cognitive Science
2005 Google
Matesius Lecture Speaker - Charles University, Prague
11
ISI/UCLA Distinguished Lecture Series
Johns Hopkins
Papers Read
1993 AAAI
1994 AAAI
6. Research in Progress
12
Eugene Charniak's research has always concerned computer
comprehension of natural language. Lately he has been concentrating
on the problem of language learning through statistical methods. The
statistical analysis of language has become more and more interesting
with the increases in computer power/memory and the increasing
availability of on-line corpora. Currently, Charniak and his students
are conducting experiments in learning probabilistic context-free
grammars (PCFG) for English from such corpora. A PCFG is a
context-free grammar which also has a probability associated with each
rule that is interpreted as the probability of the
left-hand side non-terminal being expanded using this rule (as opposed
to the other rules that could be used to expand it). There is a
standard technique for ``tuning'' PCFGs so that the associated
probabilities match the data as closely as possible. This technique,
along with restrictions on the possible form of a rule, allow one
in effect to propose all possible rules and then eliminate the ones
that after tuning have zero probability. Unfortunately, the tuning
technique only finds a locally optimal set of probabilities, and thus
one needs further information, the ``bias,'' so that the locally
optimal set is the desired set. This research is interesting both
because of the light it may shed on how much bias is actually
required and because the end product, a reasonably robust grammar for
English, is a prerequisite for further research in the area. In
particular, given a robust parsing scheme, we can now start to look at
more semantic issues like selectional restrictions on the arguments
of verbs or of adjectives. That is, we develop semantic
classes such as the things which can be the direct object of ``eat''
(presumably these would be mostly food). The end result of such
experimentation is statistical language models
that can be used to improve speech-comprehension programs, help
``stuff'' databases with information from news articles, or translate
languages.
7. Service
To the University
13
1989-1991 Committe on Nominations
To the profession
1977-1984 Editor,
Cognitive Science
14
1994 Program Committee, AAAI
Best Paper Committee, AAAI
Honors
Research Grants
15
1979-1980 NSF, ``Computer Science and Computer Engineering Research
Equipment'', $55,096.
2000-2004 NSF, "Robust Knowledge Discovery from Parallel Speech and Text"
(with large group from Brown, Johns Hopkins, etc.) $300,000.
9. Teaching
16
1993-94 Natural Language Understanding (CS241) (6)
Syllabus: http://www.cs.brown.edu/courses/cs241
2005-06 CS051
CS241
.
Theses Directed
17
1995 Glenn A. Carroll, ``Learning Probabilistic Grammars for Language
Modeling''
January, 2006
18