1 Introduction to Computational Natural Language Learning Linguistics 79400 (Under: Topics in Natural Language Processing ) Computer Science 83000 (Under:

Slides:



Advertisements
Similar presentations
Introduction: The Chomskian Perspective on Language Study.
Advertisements

Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under:
1 Language and kids Linguistics lecture #8 November 21, 2006.
MORPHOLOGY - morphemes are the building blocks that make up words.
Chapter 6 Formal Approaches to SLA Joanna – N98C0026 楊鎧綺 Gass, S. M., & Selinker, L. (2008). Second language acquisition: An introductory course (3rd.
Section 4: Language and Intelligence Overview Instructor: Sandiway Fong Department of Linguistics Department of Computer Science.
Linguistic Theory Lecture 8 Meaning and Grammar. A brief history In classical and traditional grammar not much distinction was made between grammar and.
Topics in Cognition and Language: Theory, Data and Models *Perceptual scene analysis: extraction of meaning events, causality, intentionality, Theory of.
1 Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under:
Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under:
CS 330 Programming Languages 09 / 18 / 2007 Instructor: Michael Eckmann.
Topic: Theoretical Bases for Cognitive Method Objectives Trainees will be able to give reasons for the design and procedures of the Cognitive Method.
1 Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under:
Computational Intelligence 696i Language Lecture 4 Sandiway Fong.
How can a 2-year old child decide about the underlying order in his mother’s syntax? General Problem: How do you find out about the shifting parallels.
1 CONTEXT-FREE GRAMMARS. NLE 2 Syntactic analysis (Parsing) S NPVP ATNNSVBD NP AT NNthechildrenate thecake.
Psycholinguistics 12 Language Acquisition. Three variables of language acquisition Environmental Cognitive Innate.
1 Modeling Parameter Setting Performance in Domains with a Large Number of Parameters: A Hybrid Approach CUNY / SUNY / NYU Linguistics Mini-conference.
1 Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under:
Transformational Grammar p.33 - p.43 Jack October 30 th, 2012.
Models of Generative Grammar Smriti Singh. Generative Grammar  A Generative Grammar is a set of formal rules that can generate an infinite set of sentences.
Shape of the Class: The Foundation
Lecture 1 Introduction: Linguistic Theory and Theories
Generative Grammar(Part ii)
Linguistic Theory Lecture 3 Movement. A brief history of movement Movements as ‘special rules’ proposed to capture facts that phrase structure rules cannot.
Three Generative grammars
Emergence of Syntax. Introduction  One of the most important concerns of theoretical linguistics today represents the study of the acquisition of language.
EFL Anthony’s model: Approach Method Technique
Grammars.
What counts as evidence in linguistics?. WHAT IS UNIVERSAL GRAMMAR? A system of grammatical rules and constraints believed to underlie all natural languages.
Introduction Syntax: form of a sentence (is it valid) Semantics: meaning of a sentence Valid: the frog writes neatly Invalid: swims quickly mathematics.
Evolution of Universal Grammar Pia Göser Universität Tübingen Seminar: Sprachevolution Dozent: Prof. Jäger
PS: Introduction to Psycholinguistics Winter Term 2005/06 Instructor: Daniel Wiechmann Office hours: Mon 2-3 pm Phone:
What is linguistics  It is the science of language.  Linguistics is the systematic study of language.  The field of linguistics is concerned with the.
THE BIG PICTURE Basic Assumptions Linguistics is the empirical science that studies language (or linguistic behavior) Linguistics proposes theories (models)
Psycholinguistic Theory
Unsupervised learning of Natural languages Eitan Volsky Yasmine Meroz.
Introduction to Language Acquisition Theory Janet Dean Fodor St. Petersburg July 2013 Class 1. Language acquisition theory: Origins and issues.
First Language Acquisition
1 Prof.Roseline WEEK-4 LECTURE -4 SYNTAX. 2 Prof.Roseline Syntax Concentrate on the structure and ordering of components within a sentence Greater focus.
Culture , Language and Communication
CSA2050 Introduction to Computational Linguistics Lecture 1 What is Computational Linguistics?
Center for PersonKommunikation P.1 Background for NLP Questions brought up by N. Chomsky in the 1950’ies: –Can a natural language like English be described.
Rules, Movement, Ambiguity
CSA2050 Introduction to Computational Linguistics Parsing I.
The Minimalist Program
Chapter One What is language? What is it we know about language?
Universal Grammar Functional Approaches
1Computer Sciences Department. Book: INTRODUCTION TO THE THEORY OF COMPUTATION, SECOND EDITION, by: MICHAEL SIPSER Reference 3Computer Sciences Department.
1 Syntax 1. 2 In your free time Look at the diagram again, and try to understand it. Phonetics Phonology Sounds of language Linguistics Grammar MorphologySyntax.
SYNTAX.
◦ Process of describing the structure of phrases and sentences Chapter 8 - Phrases and sentences: grammar1.
1 LIN 1310B Introduction to Linguistics Prof: Nikolay Slavkov TA: Qinghua Tang CLASS 11, Feb 9, 2007.
1 Introduction to Language Acquisition Theory Janet Dean Fodor St. Petersburg July 2013 Class 8. Implications and further questions Class 8. Implications.
Chapter 7 Linguistics aspect of interlanguage
Syntax By WJQ. Syntax : Syntax is the study of the rules governing the way words are combined to form sentences in a language, or simply, the study of.
Universal Grammar Chomsky and his followers no longer use the term LAD, but refer to the child’s innate endowment as Universal Grammar (UG). UG is a theory.
MENTAL GRAMMAR Language and mind. First half of 20 th cent. – What the main goal of linguistics should be? Behaviorism – Bloomfield: goal of linguistics.
Chapter 3 Language Acquisition: A Linguistic Treatment Jang, HaYoung Biointelligence Laborotary Seoul National University.
PSYCHOLINGUISTICS Grupo: Alejandro Ferrer Moya, Blanca Gallego Lopez,María Jose Garcia Ramón, Victoria Herrera Mercader, Soraya Lozano Carrión Grupo: Alejandro.
Ch 4. Language Acquisition: Memoryless Learning 4.1 ~ 4.3 The Computational Nature of Language Learning and Evolution Partha Niyogi 2004 Summarized by.
Syntax 1.
An Introduction to the Government and Binding Theory
2nd Language Learning Chapter 2 Lecture 4.
William Gregory Sakas City University of New York (CUNY)
Natural Language Processing
Ch 6. Language Change: Multiple Languages 6.1 Multiple Languages
Traditional Grammar VS. Generative Grammar
Linguistic aspects of interlanguage
Presentation transcript:

1 Introduction to Computational Natural Language Learning Linguistics (Under: Topics in Natural Language Processing ) Computer Science (Under: Topics in Artificial Intelligence ) The Graduate School of the City University of New York Fall 2001 William Gregory Sakas Hunter College, Department of Computer Science Graduate Center, PhD Programs in Computer Science and Linguistics The City University of New York

2 A Very Brief Linguistic History of Modern Syntax I: Phrase structure rules and transformation (Chomsky 1957, 1965) Movement and subject-aux inversion transformations: He will give a bone to the dog. He will give what to the dog. What will he give t to the dog? Base generated structure Surface structure But serious problem. The theory gives too many transformations without a good criteria for the learner to select among them. (However see Wexler and Culicover (1980) for an impressive formal characterization of how to learn transformations.)

3 A Very Brief Linguistic History of Modern Syntax II: X-bar theory (Jackendoff, 1977) Specific internal relationships between phrasal components regardless of what the "main" or head category of the phrase is. Roughly. the underlying principles that determine how the ingredients of a noun phrase, verb phrase, prepositional phrase (etc.) interact, are identical. Moreover, these principles apply cross linguistically. In English the head of a phrase (e.g. a verb or a preposition) comes before its argument (e.g. an object), whereas in Japanese, the head follows its argument.

4 A Very Brief Linguistic History of Modern Syntax II: principles and parameters (Chomsky, 1981) i.X-bar theory ii.Movement transformations (and of course the notions of base structure and surface structure) iii.Natural languages are assumed to share the same innate universal principles governing i and ii and to differ only with respect to their lexicons and the settings of a finite number of parameters. Universal Grammar (UG) = = principles and parameters. A natural language (human) grammar = = a lexicon and an array of settings (or values) of the parameters + UG.

5 principles and parameters (con't) For example: all languages have subjects of some sort Overt subjects are optional (yes / no) onoff (e.g Spanish) (e.g. English) Null Subject Parameter Language acquisition is the process of selecting the correct value of each such parameter for the language the learner is exposed to.

6 A Bit of review from first lecture: Why computationally model natural language acquisition? Pinker (1979) : "...it may be necessary to find out how language learning could work in order for the developmental data to tell us how is does work." [emphasis mine]

7 Learnability Is the learner guaranteed to converge on the target grammar for every language in a given domain? Gold (1967), Wexler and Culicover (1980), Gibson & Wexler (1994), Kanazawa (1994) An early learnability result (Gold, 1967) Exposed to input strings of an arbitrary target language generated by grammar G targ, it is impossible to guarantee that any learner can converge on G targ if G targ is drawn from any class in the Chomsky hierarchy. (E.g. context-free grammars).

8 Gold’s result is sometimes taken to be strong evidence for a nativist Universal Grammar. 1 ) Psycholinguistic research indicates that children learn grammar based on positive exemplar sentences. 2) Gold proves that G reg G cfg G cs G re can’t be learned this way. Conclude: some grammatical competence must be in place before learning commences. Gold’s result is often misapplied, but much discussion.

9 Back to new stuff: Another Learnability result: All classes of grammars possible within the principles and parameters framework are learnable because each class is of finite size. In fact a simple Blind Guess Learner is guaranteed to succeed in the long run for any finite class of grammars. But, is this not-so-blind-learner guaranteed (it has a parse test) to converge on all possible targets in a P&P domain? Error-driven Blind-Guess Learner (no oracle): 1. randomly hypothesize a current grammar 2. consume and attempt to parse a sentence from the linguistic environment 3. If the sentence is parsable by the current grammar, go to 2. Otherwise go to 1.

10 Feasibility Is acquisition possible within a reasonable amount of time and/or with a reasonable amount of work? Clark (1994, in press), Niyogi and Berwick (1996), Lightfoot (1989) (degree-0), Sakas(2000), Tesar and Smolensky (1996) and many PAC results concerning induction of FSA’s Feasibility measure (Sakas and Fodor, 2001) Near linear increase of the expected number of sentences consumed before a learner converges on the target grammar.