Intermediate depth representations

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The limitations of shallow representations have in part driven AI researchers to focus on deeper representations of knowledge. While deep representations solve some problems, they come at a computational cost. This paper focuses on the computational and representational advantages that may exist in using representations whose depth is intermediate between shallow and deep. The Roschian notion of basic level categories is used to help develop the notion of the cognitively most economic representation. For medical diagnostic systems that reason about time varying aspects of disease, it is proposed that qualitative disease histories are a good intermediate representation, lying between shallow disease patterns and deeper qualitative models. Since no single representation will provide complete coverage of a problem domain, this paper further considers how one could construct, in a principled way, a reasoning system that uses multiple representations. Measures of intra- and inter-representational adequacy are proposed to define the optimal level of such a knowledge base for a given problem. These measures define the trade-offs that occur when using a particular representational level, and the conditions under which a reasoner can decide to switch representations. As an example, the formal relationships between histories and the qualitative models that produce them are shown to define conditions that can be used by a reasoning system to switch from histories to deeper models.

LanguageEnglish
Pages431-445
Number of pages15
JournalArtificial Intelligence in Medicine
Volume4
Issue number6
DOIs
Publication statusPublished - 1992
Externally publishedYes

Fingerprint

Switches
Knowledge Bases
Economics
Research Personnel
Costs and Cost Analysis
Costs

Keywords

  • basic level
  • category theory
  • deep knowledge
  • Knowledge representation
  • multilevel model
  • qualitative disease history
  • qualitative reasoning
  • qualitative superposition

Cite this

@article{55a56a7970124a2d8a9312b6b6c97ef7,
title = "Intermediate depth representations",
abstract = "The limitations of shallow representations have in part driven AI researchers to focus on deeper representations of knowledge. While deep representations solve some problems, they come at a computational cost. This paper focuses on the computational and representational advantages that may exist in using representations whose depth is intermediate between shallow and deep. The Roschian notion of basic level categories is used to help develop the notion of the cognitively most economic representation. For medical diagnostic systems that reason about time varying aspects of disease, it is proposed that qualitative disease histories are a good intermediate representation, lying between shallow disease patterns and deeper qualitative models. Since no single representation will provide complete coverage of a problem domain, this paper further considers how one could construct, in a principled way, a reasoning system that uses multiple representations. Measures of intra- and inter-representational adequacy are proposed to define the optimal level of such a knowledge base for a given problem. These measures define the trade-offs that occur when using a particular representational level, and the conditions under which a reasoner can decide to switch representations. As an example, the formal relationships between histories and the qualitative models that produce them are shown to define conditions that can be used by a reasoning system to switch from histories to deeper models.",
keywords = "basic level, category theory, deep knowledge, Knowledge representation, multilevel model, qualitative disease history, qualitative reasoning, qualitative superposition",
author = "Enrico Coiera",
year = "1992",
doi = "10.1016/0933-3657(92)90011-D",
language = "English",
volume = "4",
pages = "431--445",
journal = "Artificial Intelligence in Medicine",
issn = "0933-3657",
publisher = "Elsevier",
number = "6",

}

Intermediate depth representations. / Coiera, Enrico.

In: Artificial Intelligence in Medicine, Vol. 4, No. 6, 1992, p. 431-445.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Intermediate depth representations

AU - Coiera, Enrico

PY - 1992

Y1 - 1992

N2 - The limitations of shallow representations have in part driven AI researchers to focus on deeper representations of knowledge. While deep representations solve some problems, they come at a computational cost. This paper focuses on the computational and representational advantages that may exist in using representations whose depth is intermediate between shallow and deep. The Roschian notion of basic level categories is used to help develop the notion of the cognitively most economic representation. For medical diagnostic systems that reason about time varying aspects of disease, it is proposed that qualitative disease histories are a good intermediate representation, lying between shallow disease patterns and deeper qualitative models. Since no single representation will provide complete coverage of a problem domain, this paper further considers how one could construct, in a principled way, a reasoning system that uses multiple representations. Measures of intra- and inter-representational adequacy are proposed to define the optimal level of such a knowledge base for a given problem. These measures define the trade-offs that occur when using a particular representational level, and the conditions under which a reasoner can decide to switch representations. As an example, the formal relationships between histories and the qualitative models that produce them are shown to define conditions that can be used by a reasoning system to switch from histories to deeper models.

AB - The limitations of shallow representations have in part driven AI researchers to focus on deeper representations of knowledge. While deep representations solve some problems, they come at a computational cost. This paper focuses on the computational and representational advantages that may exist in using representations whose depth is intermediate between shallow and deep. The Roschian notion of basic level categories is used to help develop the notion of the cognitively most economic representation. For medical diagnostic systems that reason about time varying aspects of disease, it is proposed that qualitative disease histories are a good intermediate representation, lying between shallow disease patterns and deeper qualitative models. Since no single representation will provide complete coverage of a problem domain, this paper further considers how one could construct, in a principled way, a reasoning system that uses multiple representations. Measures of intra- and inter-representational adequacy are proposed to define the optimal level of such a knowledge base for a given problem. These measures define the trade-offs that occur when using a particular representational level, and the conditions under which a reasoner can decide to switch representations. As an example, the formal relationships between histories and the qualitative models that produce them are shown to define conditions that can be used by a reasoning system to switch from histories to deeper models.

KW - basic level

KW - category theory

KW - deep knowledge

KW - Knowledge representation

KW - multilevel model

KW - qualitative disease history

KW - qualitative reasoning

KW - qualitative superposition

UR - http://www.scopus.com/inward/record.url?scp=0027064898&partnerID=8YFLogxK

U2 - 10.1016/0933-3657(92)90011-D

DO - 10.1016/0933-3657(92)90011-D

M3 - Article

VL - 4

SP - 431

EP - 445

JO - Artificial Intelligence in Medicine

T2 - Artificial Intelligence in Medicine

JF - Artificial Intelligence in Medicine

SN - 0933-3657

IS - 6

ER -