Probability and plurality for aggregations of learning machines

Leonard Pitt, Carl H. Smith

Research output: Contribution to journalArticle

Abstract

A new notion of probabilistic team inductive inference is introduced and compared with both probabilistic inference and team inference. In many cases, but not all, probabilism can be traded for pluralism, and vice versa. Necessary and sufficient conditions are given describing when a team of deterministic or probabilistic learning machines can be coalesced into a single learning machine. A subtle difference between probabilism and pluralism is revealed.

Original languageEnglish (US)
Pages (from-to)77-92
Number of pages16
JournalInformation and Computation
Volume77
Issue number1
DOIs
StatePublished - Apr 1988

Fingerprint

Learning systems
Aggregation
Machine Learning
Agglomeration
Inductive Inference
Probabilistic Inference
Necessary Conditions
Sufficient Conditions

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Information Systems
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Probability and plurality for aggregations of learning machines. / Pitt, Leonard; Smith, Carl H.

In: Information and Computation, Vol. 77, No. 1, 04.1988, p. 77-92.

Research output: Contribution to journalArticle

Pitt, Leonard ; Smith, Carl H. / Probability and plurality for aggregations of learning machines. In: Information and Computation. 1988 ; Vol. 77, No. 1. pp. 77-92.
@article{14d47e48bd414c25aed438c86437ef35,
title = "Probability and plurality for aggregations of learning machines",
abstract = "A new notion of probabilistic team inductive inference is introduced and compared with both probabilistic inference and team inference. In many cases, but not all, probabilism can be traded for pluralism, and vice versa. Necessary and sufficient conditions are given describing when a team of deterministic or probabilistic learning machines can be coalesced into a single learning machine. A subtle difference between probabilism and pluralism is revealed.",
author = "Leonard Pitt and Smith, {Carl H.}",
year = "1988",
month = "4",
doi = "10.1016/0890-5401(88)90046-6",
language = "English (US)",
volume = "77",
pages = "77--92",
journal = "Information and Computation",
issn = "0890-5401",
publisher = "Elsevier Inc.",
number = "1",

}

TY - JOUR

T1 - Probability and plurality for aggregations of learning machines

AU - Pitt, Leonard

AU - Smith, Carl H.

PY - 1988/4

Y1 - 1988/4

N2 - A new notion of probabilistic team inductive inference is introduced and compared with both probabilistic inference and team inference. In many cases, but not all, probabilism can be traded for pluralism, and vice versa. Necessary and sufficient conditions are given describing when a team of deterministic or probabilistic learning machines can be coalesced into a single learning machine. A subtle difference between probabilism and pluralism is revealed.

AB - A new notion of probabilistic team inductive inference is introduced and compared with both probabilistic inference and team inference. In many cases, but not all, probabilism can be traded for pluralism, and vice versa. Necessary and sufficient conditions are given describing when a team of deterministic or probabilistic learning machines can be coalesced into a single learning machine. A subtle difference between probabilism and pluralism is revealed.

UR - http://www.scopus.com/inward/record.url?scp=38249030604&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38249030604&partnerID=8YFLogxK

U2 - 10.1016/0890-5401(88)90046-6

DO - 10.1016/0890-5401(88)90046-6

M3 - Article

AN - SCOPUS:38249030604

VL - 77

SP - 77

EP - 92

JO - Information and Computation

JF - Information and Computation

SN - 0890-5401

IS - 1

ER -