Robot baby 2001

Paul R Cohen, Tim Oates, Niall Adams, Carole R. Beal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

In this paper we claim that meaningful representations can be learned by programs, although today they are almost always designed by skilled engineers. We discuss several kinds of meaning that representations might have, and focus on a functional notion of meaning as appropriate for programs to learn. Specifically, a representation is meaningful if it incorporates an indicator of external conditions and if the indicator relation informs action. We survey methods for inducing kinds of representations we call structural abstractions. Prototypes of sensory time series are one kind of structural abstraction, and though they are not denoting or compositional, they do support planning. Deictic representations of objects and prototype representations of words enable a program to learn the denotational meanings of words. Finally, we discuss two algorithms designed to find the macroscopic structure of episodes in a domain-independent way.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages32-56
Number of pages25
Volume2225
ISBN (Print)3540428755, 9783540428756
StatePublished - 2001
Externally publishedYes
Event12th Annual Conference on Algorithmic Learning Theory, ALT 2001 - Washington, United States
Duration: Nov 25 2001Nov 28 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2225
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other12th Annual Conference on Algorithmic Learning Theory, ALT 2001
CountryUnited States
CityWashington
Period11/25/0111/28/01

Fingerprint

Time series
Robot
Robots
Engineers
Planning
Prototype
Meaning
Abstraction
Object

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Cohen, P. R., Oates, T., Adams, N., & Beal, C. R. (2001). Robot baby 2001. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2225, pp. 32-56). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2225). Springer Verlag.

Robot baby 2001. / Cohen, Paul R; Oates, Tim; Adams, Niall; Beal, Carole R.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2225 Springer Verlag, 2001. p. 32-56 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2225).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cohen, PR, Oates, T, Adams, N & Beal, CR 2001, Robot baby 2001. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 2225, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2225, Springer Verlag, pp. 32-56, 12th Annual Conference on Algorithmic Learning Theory, ALT 2001, Washington, United States, 11/25/01.
Cohen PR, Oates T, Adams N, Beal CR. Robot baby 2001. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2225. Springer Verlag. 2001. p. 32-56. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Cohen, Paul R ; Oates, Tim ; Adams, Niall ; Beal, Carole R. / Robot baby 2001. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2225 Springer Verlag, 2001. pp. 32-56 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5334fb30017c445db2bb4d705b35e31f,
title = "Robot baby 2001",
abstract = "In this paper we claim that meaningful representations can be learned by programs, although today they are almost always designed by skilled engineers. We discuss several kinds of meaning that representations might have, and focus on a functional notion of meaning as appropriate for programs to learn. Specifically, a representation is meaningful if it incorporates an indicator of external conditions and if the indicator relation informs action. We survey methods for inducing kinds of representations we call structural abstractions. Prototypes of sensory time series are one kind of structural abstraction, and though they are not denoting or compositional, they do support planning. Deictic representations of objects and prototype representations of words enable a program to learn the denotational meanings of words. Finally, we discuss two algorithms designed to find the macroscopic structure of episodes in a domain-independent way.",
author = "Cohen, {Paul R} and Tim Oates and Niall Adams and Beal, {Carole R.}",
year = "2001",
language = "English (US)",
isbn = "3540428755",
volume = "2225",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "32--56",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Robot baby 2001

AU - Cohen, Paul R

AU - Oates, Tim

AU - Adams, Niall

AU - Beal, Carole R.

PY - 2001

Y1 - 2001

N2 - In this paper we claim that meaningful representations can be learned by programs, although today they are almost always designed by skilled engineers. We discuss several kinds of meaning that representations might have, and focus on a functional notion of meaning as appropriate for programs to learn. Specifically, a representation is meaningful if it incorporates an indicator of external conditions and if the indicator relation informs action. We survey methods for inducing kinds of representations we call structural abstractions. Prototypes of sensory time series are one kind of structural abstraction, and though they are not denoting or compositional, they do support planning. Deictic representations of objects and prototype representations of words enable a program to learn the denotational meanings of words. Finally, we discuss two algorithms designed to find the macroscopic structure of episodes in a domain-independent way.

AB - In this paper we claim that meaningful representations can be learned by programs, although today they are almost always designed by skilled engineers. We discuss several kinds of meaning that representations might have, and focus on a functional notion of meaning as appropriate for programs to learn. Specifically, a representation is meaningful if it incorporates an indicator of external conditions and if the indicator relation informs action. We survey methods for inducing kinds of representations we call structural abstractions. Prototypes of sensory time series are one kind of structural abstraction, and though they are not denoting or compositional, they do support planning. Deictic representations of objects and prototype representations of words enable a program to learn the denotational meanings of words. Finally, we discuss two algorithms designed to find the macroscopic structure of episodes in a domain-independent way.

UR - http://www.scopus.com/inward/record.url?scp=27744461891&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27744461891&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:27744461891

SN - 3540428755

SN - 9783540428756

VL - 2225

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 32

EP - 56

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -