神经网络动态分析和优化

1 BACKGROUND

1.1 Combinatorial optimization and NP-Hard problem

1.2 Traditional methods for combinatorial optimization

1.2.1 Local search

1.2.2 Genetic algorithm

1.2.3 Tabu search

1.2.4 Simulated annealing

1.3 Themes and contributions of the book

2 ARTIFICIAL NEURAL NETWORK

2.1 Biological neural network to artificial neural network

2.2 History of artificial neural network

2.3 Methods of neural networks

2.3.1 Hopfield neural network

2.3.2 Self-organizing map

2.3.3 Maximum neural network

2.3.4 Elastic net

2.4 Problems of neural network

3 IMPROVED HOPFIELD-TYPE NN WITH CHAOTIC DYNAMICS FOR MCP

3.1 Maximum clique problem

3.2 The analysis of HNN with chaotic dynamics

3.2.1 Chaotic neural networks

3.2.2 Transiently chaotic neural network

3.2.3 Relationship between TCNN and CNN

3.2.4 Analysis of parameter selection

3.3 A flexible TCNN for MCP

3.3.1 The flexible annealing strategy

3.3.2 Flexible TCNN and its dynamic analysis

3.3.3 Simulations

3.4 TCNN with filter method for MCP

3.4.1 Analysis on feasibility and adaptivity of TCNN

3.4.2 Algorithm description

3.4.3 Simulations

3.5 Delayed TCNN and its application on MCP

3.5.1 The flaw of variable delayed model

3.5.2 The delayed transiently chaotic neural network

3.5.3 Simulation

3.6 Summary

4 IMPROVED CHAOTIC MNN FOR COPS

4.1 Chaotic maximum neural network

4.2 Improved CMNN with stochastic dynamics for N-Queens

4.2.1 N-Queens problems

4.2.2 Dynamics analysis and Improvement of the CMNN with

stochastic dynamics for N-Queens problems

4.2.3 Algorithm description

4.2.4 Simulations

4.3 Chaotic MNN with flexible annealing strategy for MCP

4.3.1 Improved algorithm and its dynamics analysis

4.3.2 Simulations

4.4 Summary

5 IMPROVED ELASTIC NET FOR TRAVELING SAI-ESMAN PROBLEM

5.1 Elastic net for TSP

5.2 Efficiency analysis on elastic net

5.3 The improved algorithms based on Elastic Net

5.3.1 Rebuilt clone elastic net algorithm

5.3.2 The unsupervised up-to-bottom hierarchical clustering elas.

tic net algorithm

5.3.3 Simulation

5.4 Summary

6 SUMMARY AND FUTURE WORK

6.1 Summary of this book

6.2 Future work

BIBLIOGRAPHY

《神经网络动态分析和优化（英文版）》以霍普菲尔德网络和弹性网络为重点，集中介绍了神经网络算法的动态特性和控制方法，显示了神经网络复杂动态对网络性能的重要影响；并在神经网络动态分析的基础上，给出了优化神经网络的有效方法和核心理论。《神经网络动态分析和优化（英文版）》的诸多方法和理论，展现了神经网络算法在组合优化问题上的优势，为神经网络算法的优化和应用，提供了启发和指导思想；并且，《神经网络动态分析和优化（英文版）》进一步完善了神经网络基础理论，具有重要的理论和应用价值。