## Abstract

A unified approach is developed for hard optimization problems involving data association, i.e. the assignment of elements viewed as 'data' to one of a set classes so as to minimize the resulting cost. The diverse problems which fit this description include data clustering, statistical classifier design to minimize probability of error, piece-wise regression structure vector quantization, as well as optimization in graph theory. Whereas standard descent-based methods are susceptible to finding poor local optima of the cost, the suggested approach provides some potential for avoiding local optima, yet without the computational complexity of stochastic annealing.

Original language | English (US) |
---|---|

Number of pages | 1 |

State | Published - Jan 1 1995 |

Event | Proceedings of the 1995 IEEE International Symposium on Information Theory - Whistler, BC, Can Duration: Sep 17 1995 → Sep 22 1995 |

### Other

Other | Proceedings of the 1995 IEEE International Symposium on Information Theory |
---|---|

City | Whistler, BC, Can |

Period | 9/17/95 → 9/22/95 |

## All Science Journal Classification (ASJC) codes

- Theoretical Computer Science
- Information Systems
- Modeling and Simulation
- Applied Mathematics