CSC2540S Machine Learning and Universal Grammar
Department of Computer Science,
Tuesday
Shalom Lappin
Department of Philosophy
King's College
shalom@cs.toronto.edu
Please note that the class
has moved from MS3268 to SF3207
Recent advances in the
application of machine learning to grammar induction and related
NLP tasks have shown that it
is possible to acquire a significant amount of linguistic
knowledge through domain general learning algorithms
constrained by relatively weak
biased language models. Computational learning theory has
also provided important
results for understanding the limits and possibilities of
different sorts of learning
algorithms for acquiring the grammars of formal and natural
languages. The seminar
considers the implications of this work for human language
acquisition and the theory
of universal grammar, the language model that defines
the space of possible hypotheses
for natural language grammars. While the seminar focuses
on research in machine
learning and computational learning theory, it will also look
at some current work in
theoretical linguistics, psycholinguistics, cognitive science,
and genetics in exploring
these issues.
Week 1:
Overview
of the problem,
The
argument from the poverty of stimulus (APS) and Universal Grammar
Week 2: January 13, Cognitive
Architecture and Language Evolution
Nativism and cognitive architecture,
The
evolution of language
Week 3: January 20, Clarifying
the APS
A
critical examination of some linguistic arguments for the APS
Week 4: January 27, The Nature of Primary Linguistic Evidence,
The role
of negative evidence in language acquisition
Week 5: February 3, Identification
in the Limit
The Gold
model of grammar induction
Week 6: February 10, PAC
Learning
An
alternative computational learning paradigm
Week 7: February 17, Reading Week
Week 8: February 24, Modifying
PAC Learning with Distributions
Using
probability distributions on learning samples to facilitate grammar induction
Week 9: March 3, Machine
Learning and Grammar Induction
Unsupervised
grammar induction,
Recent
work on acquiring grammars through the application of machine learning
methods to large natural language corpora
Week 10: March 10, Parameters
in Linguistic Theory and in Probabilistic Language
Models
Parametric
explanations of language universals and variation in linguistic theory,
A bootstrapping approach to parameter
estimation for language models
Week 11: March 17, Genetic
and Psycholinguistic Evidence
The FOXP2
gene and specific language impairment,
Psychological
evidence for Bayesian grammar induction
Weeks 12-14: March 24, March 31, and April 7, Project
Presentations