A tutorial on the cross-entropy method

被引：2101

作者：

De Boer, PT

Kroese, DP

Mannor, S

Rubinstein, RY

机构：

[1] Univ Twente, Dept Elect Engn Math & Comp Sci, NL-7500 AE Enschede, Netherlands

[2] Univ Queensland, Dept Math, Brisbane, Qld 4072, Australia

[3] McGill Univ, Dept Elect & Comp Engn, Montreal, PQ, Canada

[4] Technion Israel Inst Technol, Dept Ind Engn, IL-32000 Haifa, Israel

来源：

ANNALS OF OPERATIONS RESEARCH | 2005年 / 134卷 / 01期

基金：

以色列科学基金会;

关键词：

cross-entropy method; Monte-Carlo simulation; randomized optimization; machine learning; rare events;

D O I：

10.1007/s10479-005-5724-z

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

The cross-entropy (CE) method is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the CE method. We present the CE methodology, the basic algorithm and its modifications, and discuss applications in combinatorial optimization and machine learning.

引用

页码：19 / 67

页数：49

共 50 条

[1] Aarts E., 1989, Wiley-Interscience Series in Discrete Mathematics and Optimization
[2] Alon G., 2005, ANN OPER RES, V134, P19
[3] [Anonymous], 1989, GENETIC ALGORITHM SE
[4] [Anonymous], 1979, Computers and Intractablity: A Guide to the Theoryof NP-Completeness
[5] ASMUSSEN S, 2005, IN PRESS STOCHASTIC, V21
[6] NEURONLIKE ADAPTIVE ELEMENTS THAT CAN SOLVE DIFFICULT LEARNING CONTROL-PROBLEMS
BARTO, AG
SUTTON, RS
ANDERSON, CW
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1983, 13 (05): : 834 - 846
[7] BARTON AG, 1998, REINFORCEMENT LEARNI
[8] Experiments with infinite-horizon, policy-gradient estimation
Baxter, J
Bartlett, PL
Weaver, L
[J]. JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2001, 15 : 351 - 381
[9] Bertsekas DP, 2012, DYNAMIC PROGRAMMING, V2
[10] BERTSEKAS DP, 1995, NEURODYNAMIC PROGRAM

← 1 2 3 4 5 →