skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A simple way to improve path consistency processing in interval algebra networks

Abstract

Reasoning about qualitative temporal information is essential in many artificial intelligence problems. In particular, many tasks can be solved using the interval-based temporal algebra introduced by Allen (A1183). In this framework, one of the main tasks is to compute the transitive closure of a network of relations between intervals (also called path consistency in a CSP-like terminology). Almost all previous path consistency algorithms proposed in the temporal reasoning literature were based on the constraint reasoning algorithms PC-1 and PC-2 (Mac77). In this paper, we first show that the most efficient of these algorithms is the one which stays the closest to PC-2. Afterwards, we propose a new algorithm, using the idea {open_quotes}one support is sufficient{close_quotes} (as AC-3 (Mac77) does for arc consistency in constraint networks). Actually, to apply this idea, we simply changed the way composition-intersection of relations was achieved during the path consistency process in previous algorithms.

Authors:
 [1]
  1. LIRMM, Montpellier (France)
Publication Date:
OSTI Identifier:
430683
Report Number(s):
CONF-960876-
TRN: 96:006521-0058
Resource Type:
Conference
Resource Relation:
Conference: 13. National conference on artifical intelligence and the 8. Innovative applications of artificial intelligence conference, Portland, OR (United States), 4-8 Aug 1996; Other Information: PBD: 1996; Related Information: Is Part Of Proceedings of the thirteenth national conference on artificial intelligence and the eighth innovative applications of artificial intelligence conference. Volume 1 and 2; PB: 1626 p.
Country of Publication:
United States
Language:
English
Subject:
99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; ARTIFICIAL INTELLIGENCE; ALGORITHMS; NETWORK ANALYSIS; DECISION MAKING

Citation Formats

Bessiere, C. A simple way to improve path consistency processing in interval algebra networks. United States: N. p., 1996. Web.
Bessiere, C. A simple way to improve path consistency processing in interval algebra networks. United States.
Bessiere, C. Tue . "A simple way to improve path consistency processing in interval algebra networks". United States. doi:.
@article{osti_430683,
title = {A simple way to improve path consistency processing in interval algebra networks},
author = {Bessiere, C.},
abstractNote = {Reasoning about qualitative temporal information is essential in many artificial intelligence problems. In particular, many tasks can be solved using the interval-based temporal algebra introduced by Allen (A1183). In this framework, one of the main tasks is to compute the transitive closure of a network of relations between intervals (also called path consistency in a CSP-like terminology). Almost all previous path consistency algorithms proposed in the temporal reasoning literature were based on the constraint reasoning algorithms PC-1 and PC-2 (Mac77). In this paper, we first show that the most efficient of these algorithms is the one which stays the closest to PC-2. Afterwards, we propose a new algorithm, using the idea {open_quotes}one support is sufficient{close_quotes} (as AC-3 (Mac77) does for arc consistency in constraint networks). Actually, to apply this idea, we simply changed the way composition-intersection of relations was achieved during the path consistency process in previous algorithms.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Dec 31 00:00:00 EST 1996},
month = {Tue Dec 31 00:00:00 EST 1996}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • This paper continues Nebel and Burckert`s investigation of Allen`s interval algebra by presenting nine more maximal tractable subclasses of the algebra (provided that P {ne} NP), in addition to their previously reported ORD-Horn subclass. Furthermore, twelve tractable subclasses are identified, whose maximality is riot decided. Four of these can express the notion of sequentiality between intervals, which is not possible in the ORD-Horn algebra. The satisfiability algorithm, which is common for all the algebras, is shown to be linear.
  • Within the framework of constraint programming, particulary concerning the Constraint Satisfaction Problems (CSPs), the techniques of preprocessing based on filtering algorithms were shown to be very important for the search phase. In particular, two filtering methods have been studied, these methods exploit two properties of local consistency: arc- and path-consistency. Concerning the arc-consistency methods, there is a linear time algorithm (in the size of the problem) which is efficient in practice. But the limitations of the arc-consistency algorithms requires often filtering methods with higher order like path-consistency filterings. The best path-consistency algorithm proposed is PC-6, a natural generalization of AC-6more » to path-consistency. Its time complexity is O(n{sup 3}d{sup 4}) and its space complexity is O(n{sup 3}d{sup 4}), where n is the number of variables and d is the size of domains. We have remarked that PC-6, though it is widely better than PC-4, was not very efficient in practice, specially for those classes of problems that require an important space to be run. Therefore, we propose here a new path-consistency algorithm called PC-7, its space complexity is O(n{sup 3}d{sup 4}) but its time complexity is O(n{sup 3}d{sup 4}) i.e. worse than that of PC-6. However, the simplicity of PC-7 as well as the data structures used for its implementation offer really a higher performance than PC-6. Furthermore, it turns out that when the size of domains is a constant of the problems, the time complexity of PC-7 becomes. like PC-6, optimal i.e. O(n{sup 3}).« less
  • Traditional neural networks like multi-layered perceptrons (MLP) use example patterns, i.e., pairs of real-valued observation vectors, ({rvec x},{rvec y}), to approximate function {cflx f}({rvec x}) = {rvec y}. To determine the parameters of the approximation, a special version of the gradient descent method called back-propagation is widely used. In many situations, observations of the input and output variables are not precise; instead, we usually have intervals of possible values. The imprecision could be due to the limited accuracy of the measuring instrument or could reflect genuine uncertainty in the observed variables. In such situation input and output data consist ofmore » mixed data types; intervals and precise numbers. Function approximation in interval domains is considered in this paper. We discuss a modification of the classical backpropagation learning algorithm to interval domains. Results are presented with simple examples demonstrating few properties of nonlinear interval mapping as noise resistance and finding set of solutions to the function approximation problem.« less
  • The locking efficiency and stability of a Lambda Physik EMG150 tunable ArF laser was significantly improved in a simple way. The active gain volume of the oscillator laser was increased by a factor of 5 by replacing the 1-mm circular holes supplied with the original laser by 1 mm x 5 mm rectangular apertures. The resulting rectangular output of the oscillator was condensed with a cylindrical telescope and used for injection locking. The scanning range, locking efficiency, and stability of the laser were significantly improved.