Neurons or Symbols

Why does OR Remain Exclusive?

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Neuro-Symbolic Integration is an interdisciplinary area that endeavours to unify neural networks and symbolic logic. The goal is to create a system that combines the advantages of neural networks (adaptive behaviour, robustness, tolerance of noise and probability) and symbolic logic (validity of computations, generality, higher-order reasoning). Several different approaches have been proposed in the past. However, the existing neuro-symbolic networks provide only a limited coverage of the techniques used in computational logic. In this paper, we outline the areas of neuro-symbolism where computational logic has been implemented so far, and analyse the problematic areas. We show why certain concepts cannot be implemented using the existing neuro-symbolic networks, and propose four main improvements needed to build neuro-symbolic networks of the future.

Original languageEnglish
Title of host publicationIJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009
EditorsAntónio Dourado Correia, Agostinho C. Rosa, Kurosh Madani
PublisherINSTICC Press
Pages502-507
Number of pages6
ISBN (Print)9789896740146
Publication statusPublished - 2009
EventInternational Joint Conference on Computational Intelligence 2009 - Funchal, Portugal
Duration: 5 Oct 20097 Oct 2009

Conference

ConferenceInternational Joint Conference on Computational Intelligence 2009
Abbreviated titleIJCCI 2009
CountryPortugal
CityFunchal
Period5/10/097/10/09

Fingerprint

Neurons
Neural networks

Keywords

  • Computational Logic in Neural Networks
  • Neuro-Symbolic Networks
  • Connectionism
  • Hybrid networks

Cite this

Komendantskaya, E. (2009). Neurons or Symbols: Why does OR Remain Exclusive? In A. Dourado Correia, A. C. Rosa, & K. Madani (Eds.), IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009 (pp. 502-507). INSTICC Press .
Komendantskaya, Ekaterina. / Neurons or Symbols : Why does OR Remain Exclusive?. IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009. editor / António Dourado Correia ; Agostinho C. Rosa ; Kurosh Madani. INSTICC Press , 2009. pp. 502-507
@inbook{1da46d1fd30c4b11a0515f57521ddaa5,
title = "Neurons or Symbols: Why does OR Remain Exclusive?",
abstract = "Neuro-Symbolic Integration is an interdisciplinary area that endeavours to unify neural networks and symbolic logic. The goal is to create a system that combines the advantages of neural networks (adaptive behaviour, robustness, tolerance of noise and probability) and symbolic logic (validity of computations, generality, higher-order reasoning). Several different approaches have been proposed in the past. However, the existing neuro-symbolic networks provide only a limited coverage of the techniques used in computational logic. In this paper, we outline the areas of neuro-symbolism where computational logic has been implemented so far, and analyse the problematic areas. We show why certain concepts cannot be implemented using the existing neuro-symbolic networks, and propose four main improvements needed to build neuro-symbolic networks of the future.",
keywords = "Computational Logic in Neural Networks, Neuro-Symbolic Networks, Connectionism, Hybrid networks",
author = "Ekaterina Komendantskaya",
year = "2009",
language = "English",
isbn = "9789896740146",
pages = "502--507",
editor = "{Dourado Correia}, Ant{\'o}nio and Rosa, {Agostinho C.} and Kurosh Madani",
booktitle = "IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009",
publisher = "INSTICC Press",

}

Komendantskaya, E 2009, Neurons or Symbols: Why does OR Remain Exclusive? in A Dourado Correia, AC Rosa & K Madani (eds), IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009. INSTICC Press , pp. 502-507, International Joint Conference on Computational Intelligence 2009, Funchal, Portugal, 5/10/09.

Neurons or Symbols : Why does OR Remain Exclusive? / Komendantskaya, Ekaterina.

IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009. ed. / António Dourado Correia; Agostinho C. Rosa; Kurosh Madani. INSTICC Press , 2009. p. 502-507.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Neurons or Symbols

T2 - Why does OR Remain Exclusive?

AU - Komendantskaya, Ekaterina

PY - 2009

Y1 - 2009

N2 - Neuro-Symbolic Integration is an interdisciplinary area that endeavours to unify neural networks and symbolic logic. The goal is to create a system that combines the advantages of neural networks (adaptive behaviour, robustness, tolerance of noise and probability) and symbolic logic (validity of computations, generality, higher-order reasoning). Several different approaches have been proposed in the past. However, the existing neuro-symbolic networks provide only a limited coverage of the techniques used in computational logic. In this paper, we outline the areas of neuro-symbolism where computational logic has been implemented so far, and analyse the problematic areas. We show why certain concepts cannot be implemented using the existing neuro-symbolic networks, and propose four main improvements needed to build neuro-symbolic networks of the future.

AB - Neuro-Symbolic Integration is an interdisciplinary area that endeavours to unify neural networks and symbolic logic. The goal is to create a system that combines the advantages of neural networks (adaptive behaviour, robustness, tolerance of noise and probability) and symbolic logic (validity of computations, generality, higher-order reasoning). Several different approaches have been proposed in the past. However, the existing neuro-symbolic networks provide only a limited coverage of the techniques used in computational logic. In this paper, we outline the areas of neuro-symbolism where computational logic has been implemented so far, and analyse the problematic areas. We show why certain concepts cannot be implemented using the existing neuro-symbolic networks, and propose four main improvements needed to build neuro-symbolic networks of the future.

KW - Computational Logic in Neural Networks

KW - Neuro-Symbolic Networks

KW - Connectionism

KW - Hybrid networks

M3 - Chapter

SN - 9789896740146

SP - 502

EP - 507

BT - IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009

A2 - Dourado Correia, António

A2 - Rosa, Agostinho C.

A2 - Madani, Kurosh

PB - INSTICC Press

ER -

Komendantskaya E. Neurons or Symbols: Why does OR Remain Exclusive? In Dourado Correia A, Rosa AC, Madani K, editors, IJCCI 2009 - Proceedings of the International Joint Conference on Computational Intelligence, Funchal, Madeira, Portugal, October 5-7, 2009. INSTICC Press . 2009. p. 502-507