退火類神經網路融合模擬退火與霍普菲爾坦克神經網路的特徵，不但具有模擬退火演算法的高解答品質，且具有與霍普菲爾網路快速收斂的優點。而行動代理者規劃為資訊擷取系統中重要的技術之一，可在行動計算環境中提供最小成本的位置感知之服務。本研究考慮有效資源的時間限制以及行動代理者成本最佳化的探討，我們改良退火類神經網路，設計了一個新的能量函數並控制退火溫度，處理計算環境之動態時間特色和安排行動代理者行程時之伺服器效能和網路延遲；而且也滿足基於位置之限制，如旅行行動代理者行程之起訖點必為其本地端網站；轉換能量函數為李亞普諾夫函數形式結合退火溫度設計其活化函數以確保收斂至穩定狀態和合法解的存在，並設計神經元間之連結加權值計算方式和在動態網路中狀態變數之活化函數以搜尋合法解。模擬並實驗不同變數，如時間變數、退火溫度和能量函數參數值，以評估提出之模型和演算法；且以田口實驗法決定最佳參數組合。實驗結果顯示，改良後之退火類神經網路呈現模擬退火和霍普菲爾網路兩者之特徵，可快速收斂且具高解答品質；且在多節點的行動代理者問題上，更能突顯本研究設計方法的優勢。本創新之方法，可望有效善最佳化問題之效率。

Annealed neural network combines the characteristics of both simulation annealing and Hopfield-Tank neural network, which are high quality solutions and fast convergence. Mobile agent planning is an important technique of information retrieval systems to provide the minimum cost of the location-aware services in mobile computing environment. By taking the time constraints of effective resources into account and the mobile agent to explore the cost optimization, we modify annealing neural network to design a new energy function and control the annealing temperature in order to deal with the dynamic temporal feature of computing environments. We not only consider the server performance and network latency when scheduling mobile agents, but also investigate the location-based constraints, such as the home site of routing sequence of the traveling mobile agent must be the start and end node. To guarantee the convergent stable state and existence of the valid solution, the energy function is reformulated into a Lyapunov function which is combined with the annealing temperature to form an activation function. The connection weights between the neurons and the activation function of state variables in the dynamic network are devised in searching for the valid solutions. Simulation of different coefficients assess the proposed model and algorithm. Furthermore, Taguchi method is used to obtain the optimal combination factors of annealing neural network. The results show that this research presents the feature of both simulated annealing and Hopfield neural network by providing fast convergence and highly quality. In addition with a larger number of sites, the experimental results demonstrate the benefits of the annealed neural network. This innovation would be applicable to improve the effectiveness of solving optimization problems.

CHAPTER 1 INTRODUCTION......................................................................................1
CHAPTER 2 LITERATURE REVIEW............................................................................5
2.1 THEMOBILEAGENT.................................................................................................5
2.2 THEMAPPROBLEM..................................................................................................7
2.3 ARTIFICIALNEURALNETWORK............................................................................8
2.4 HOPFIELD-TANK NEURALNETWORK................................................................11
2.5 SIMULATEDANNEALING .......................................................................................14
2.6 ANNEALED NEURALNETWORK..........................................................................16
2.7 TAGUCHI METHOD.................................................................................................19
CHAPTER 3 THE MOBILE AGENT PLANNING PROBLEM...................................21
CHAPTER 4 THE MOBILE AGENT PLANNINGMODEL...........................................27
4.1 THE STATE VARIABLES.........................................................................................28
4.2 THE CONSTRAINTS AND PROBLEM GOAL .....................................................29
4.3 THEMAPENERGY FUNCTION...............................................................................38
4.4 THE CONNECTIONWEIGHTMATRICES .............................................................45
4.5 THEACTIVATION FUNCTION.................................................................................49
4.6 THEMAPALGORITHM...............................................................................................51
CHAPTER 5 SIMULATION RESULTS..........................................................................53
5.1 SIMULATION DESIGN..............................................................................................53
5.2 ENERGYCOEFFICIENTS .......................................................................................56
5.2.1 The Iteration Values .............................................................................................57
5.2.2 The Coefficient A....................................................................................................59
5.2.3 The Coefficient D...................................................................................................61
5.2.4 The Coefficient F ...................................................................................................62
5.3 THE INITIAL STATE...................................................................................................64
5.3.1 The Initial Temperature ........................................................................................64
5.3.2 The Temperature Cooling Rate..........................................................................67
5.4 SIMULATION OF 30-SITEMAPPROBLEM.............................................................69
5.5 THE COMPARISONWITH HOPFIELD-TANK NEURALNETWORK..................71
5.5.1 10-site MAP Problem ...........................................................................................71
5.5.2 Parametric Optimization of ANN Process for 10-site MAP Problem............73
5.5.3 30-site MAP Problem ...........................................................................................77
CHAPTER 6 CONCLUSION .........................................................................................80
REFERENCES ................................................................................................................82
APPENDIX I.......................................................................................................................85
APPENDIX II .....................................................................................................................88

[1] 王進德，蕭大全，類神經網路與模糊控制理論入門，全華科技圖書股份有限公司，September 2003。
[2] 葉怡成，類神經網路模式應用與實作，儒林圖書有限公司，October 2004。
[3] 吳宗益，應用基因演算法及田口實驗法於模具生產排程系統之研究，國立高雄第一科技大學碩士論文，2002。
[4] 李成禎，應用田口法實現感應電動機之參數量測與強健速度控制器設計，中正理工學院碩士論文，1998。
[5] J. Baek, J. Yeo, G. kim, and H. Yeom, “Cost effective mobile agent planning for distributed information retrieval,” in Proceedings of the 21st International Conference on Distributed Computing Systems, pp. 65 – 72, April 2001.
[6] J. Baek, G. kim, and H. Yeom, “Cost effective planning of timed mobile agent,” in Proceedings of the International Conference on Information Technology: Coding and Computing, pp. 536 – 541, April 2002.
[7] J. Baek, J. Yeo, and H. Yeom, “Agent chaining: An approach to dynamic mobile agent planning,” in Proceedings of the 22nd International Conference on Distributed Computing Systems, pp.579 – 586, July 2002.
[8] J. Baek and H. Yeom, “d-Agent: an approach to mobile agent planning for distributed information retrieval,” IEEE Transactions on Consumer Electronics, Vol. 49, Issue 1, pp. 115 – 122, January 2003.
[9] M. Carey and D. Johnson, “Computers and Intractability: A Guide to the Theory of NP-Completeness,” Freeman, San Francisco, 1979.
[10] D. Chess, C. Harrison, and A. Kershenbaum, ” Mobile Agents：Are They a Good Idea?”, IBM Research Report, March 1995.
[11] R. M. Chen, S. T. Lo, and Y. M. Huang, “Combining competitive scheme with slack neurons to solve real-time job scheduling problem,“ Expert Systems with Applications, Vol. 33, pp. 75-85, 2007.
[12] S. Datta, A. Bandyopadhyay, P. K. Pal, " Grey-based taguchi method for optimization of bead geometry in submerged arc bead-on-plate welding,” The International Journal of Advanced Manufacturing Technology, Vol. 39, 2008.
[13] E. V. David and T. K. Miller, “A traveling salesperson objective function that works,” IEEE International Conference on Neural Networks, Vol. 2, pp. 299-303, July 1988.
[14] A. Dekkers and E. Aarts, “Global optimization and simulated annealing,” Mathematical Programming, Vol. 50, pp. 367-393, 1991.
[15] A. Fuggetta, G.P. Picco, and G. Vigna, “Understanding Code Mobility,” IEEE Transactions on Software Engineering, Vol. 24, No. 5, May 1998.
[16] J. J. Hopfield, “Neural networks and physical systems with emergent collective computational abilities,” in Proceedings of the National Academy of Sciences, Vol. 79, pp. 2554-2558, April 1982.
[17] J. J. Hopfield and D. W. Tank, “'Neural' computation of decisions in optimization problems,” Biological Cybernetics, Vol. 52, pp. 141 - 152, 1985.
[18] S. Kirkpatric, C. D. Gelatt, Jr., and M. P. Vecchi, “Optimization by Simulated Annealing,” Science, Vol. 220, No. 4598, pp. 671-680, May 1983.
[19] S. Kirkpatrick, “Optimization by Simulated Annealing: Quantitative Studies,” Journal of Statistical Physics, Vol. 34, Nos. 5/6, 1984.
[20] D. B. Lange and M. Oshima, “Seven Good Reasons for Mobile Agents: Dispatch Your Agents, Shut off Your Machine,” Communications of the ACM, Vol. 42, No. 3, pp. 88-89, March 1999.
[21] C. H. Lin and J. F. Wang, “Solving the Mobile Agent Planning Problem with a Hopfield-Tank Neural Network,” in Proceedings of the International MultiConference of Engineers and Computer Scientists 2006, pp. 104-114, 20-22 June, 2006.
[22] X. D. Liu, X. Wand, Z. Wu, and X. Gu, ”Modified Hopfield Network for CDMA Multiuser Detection,” ISNN 2006, LNCS 3973, pp. 88-93, 2006.
[23] K. Moizumi, ”Mobile Agent Planning Problems,” PhD thesis, Thayer School of Engineering, Dartmouth College, November 1998.
[24] M. Takeda and J. W. Goodman, “Neural networks for computation: number representations and programming complexity,” Applied Optics, Vol. 25, No. 18, 15 September 1986.
[25] G. Taguchi, “Introduction to Quality Engineering,” Asian Productivity Organization, Tokyo, 1990.
[26] D. E. Van den Bout and T. K. Miller, “A traveling salesman objective function that works,” Proceedings of International Joint Conference on Neural Networks 1989, Washington, DC, 1989, pp. 299– 303.
[27] D. E. Van den Bout and T. K. Miller, “Graph partitioning using annealed neural networks,“ IEEE Transaction on Neural Networks, Vol.1, No. 4, pp. 192–203, 1990.
[28] I. C. Yeh and C. M. Li, “Architecture space layout using annealed neural network,” Journal of Technology, Vol. 20, No. 4,pp. 367-376, 2005.
[29] I. C. Yeh, “Architectural layout optimization using annealed neural network,” Automation in Construction, Vol. 15, pp. 531-539, 2006.