Connectionist-network auction algorithm for the assignment problem

Author

Seow, S.M. ; Lam, K.P.

Author_Institution

Dept. of Syst. Eng. & Eng. Manage., Chinese Univ. of Hong Kong, Shatin, Hong Kong

Volume

3

fYear

1995

fDate

Nov/Dec 1995

Firstpage

1225

Abstract

The assignment problem has been much studied in AI and operations research, and the auction algorithm is a well-known method for solving the problem. While several parallel computational schemes have been reported, the authors propose an alternative connectionist approach based on the binary relation inference network, which provides a unified framework for discrete-time and continuous-time systems design. In this paper, synchronous and asynchronous implementations of this connectionist-network auction algorithm are presented. Issues on the architecture and algorithm of the implementation are discussed

Keywords

combinatorial mathematics; computational complexity; inference mechanisms; neural nets; operations research; optimisation; assignment problem; asynchronous implementation; binary relation inference network; connectionist-network auction algorithm; continuous-time systems; discrete-time systems; synchronous implementation; Artificial intelligence; Computer architecture; Computer networks; Concurrent computing; Inference algorithms; Operations research; Research and development management; Shortest path problem; Systems engineering and theory; Traveling salesman problems;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1995. Proceedings., IEEE International Conference on

Conference_Location

Perth, WA

Print_ISBN

0-7803-2768-3

Type

conf

DOI

10.1109/ICNN.1995.487329

Filename

487329