Formal Learning TheoryCorrespondence to D. Osherson, DIPSCO, Istituto San Rafaelle, Via Olgettina 60, I-20132 Milano, Italy

Daniel Osherson, Dick de Jongh, Eric Martin, Scott Weinstein

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

This chapter is devoted to formal models of language acquisition, and of empirical inquiry more generally. Central to the theory is the concept of a paradigm (or model) of empirical inquiry. The inquiry in question might be that of a child learning language, or of a scientist investigating nature. Every paradigm in the theory has essentially the same stock of component concepts. A paradigm offers formal reconstruction of the following concepts: a theoretically possible reality, an intelligible hypothesis about reality, the data available about any given reality, a scientist (or child), and successful behavior by a scientist working in a given, possible reality. Different paradigms formalize this picture in different ways, resulting in different games. Whether a particular game is winnable depends, among other things, on the breadth of the set of possible realities. Wider sets make successful learning more difficult, to the point of impossibility. The dominant concern of learning theory is to formulate an illuminating characterization of the paradigms in which success is achievable. There is no better introduction to learning theory than presentation of its most fundamental paradigm.

Original languageEnglish (US)
Title of host publicationHandbook of Logic and Language
PublisherElsevier
Pages725-763
Number of pages39
ISBN (Print)9780444537263
DOIs
StatePublished - 2011

All Science Journal Classification (ASJC) codes

  • General Engineering

Keywords

  • Formal Learning Theory
  • Language Acquisition
  • Machine Inductive Inference

Fingerprint

Dive into the research topics of 'Formal Learning TheoryCorrespondence to D. Osherson, DIPSCO, Istituto San Rafaelle, Via Olgettina 60, I-20132 Milano, Italy'. Together they form a unique fingerprint.

Cite this