隨著「連續值局限型波茲曼模型(Continuous Restricted Boltzmann Machine)」神經網路系統在積體電路上的實現，如何將此網路神經系統實際應用於自然界中的實驗與測試，將是下一個重要的發展方向，由於從自然界中所取得的訊號是複雜而且具有較高之維度，所以如何運用「連續值局限型波茲曼模型」神經網路系統來處理更高階且非線性的資料是目前所需解決的問題，若能解決此問題將可使整個網路神經系統更具可行性。
本人之研究主要目的在研發一具機率行為且可自動調變的晶片系統，作為植入式(implantable)或實驗室晶片化(Lab-on-a-chip)生醫檢測系統中的智慧型訊號辨識。基於先前研究者研發「連續值局限型波茲曼模型」模型的理論與雛型(prototype)晶片的經驗，本計畫將實現一可程式化且模組化的「連續值局限型波茲曼模型」晶片系統，可程式化參數的設計以Matlab模擬找出所需之最小位元數，而模組化設計則使「連續值局限型波茲曼模型」系統可以視輸入資料來增加系統晶片個數，以達到擴大神經網路系統之功能，如此即可使其在生醫檢測方面能有更多的應用。
機率型類神經網路模型能應用隨機性來概括訊號中的變異，目前且已廣泛應用於許多生醫訊號分析，所謂的機率型即表示輸入資料僅僅決定輸出值之機率分佈，而不直接決定輸出值，其輸出值再根據此決定的機率取樣而得。故當輸入值不變時，輸出值依然可能不同，且輸出值是有特定的機率分布。機率行類神經網路模型雖然已經實際運用積體電路模式表現出來，但是在此論文的研究部分便是將此系統加以模組化以增加神經網路系統應用於實際生醫訊號檢測上的可行性。

參考文獻

【1】 葉怡成"類神經網路模式應用與實作"儒林圖書有限公司

【2】 H. Chen, "Continuous-valued Probabilistic Neural Computation in VLSI", PhD Thesis, Edinburgh University, UK, 2004

【3】 Morie, T.; Amemiya, Y.; "An all-analog expandable neural network LSI with on-chip backpropagation learning" IEEE Journal of Solid-State Circuits , Volume 29 Issue 9 Sept. 1994 Page(s):1086 – 1093

【4】 W.A.; Fujimoto, R.J.; Okamura, M.M.; "The Lockheed programmable analog neural network processor Fisher" Neural Networks, 1990., 1990 IJCNN International Joint Conference on 17-21 June 1990 Page(s):563 - 568 vol.2

【5】 C. Lu,;B. Shi,;L. Chen,;" A programmable on-chip BP learning neural network with enhanced neuron characteristics"Circuits and Systems, 2001. ISCAS 2001. The 2001 IEEE International Symposium on Volume 3, 6-9 May 2001 Page(s):573 - 576 vol. 2

【6】 J. J. Hopfield, "Neural networks and physical systems with emergent collective computationalabilities,". Proceedings of the National Academy of Sciences of the U.S.A., vol. 79,pp. 2554.2558, 1982.

【7】 Thiam, T.; "The Boltzmann machine"Neural Networks, 1999. IJCNN '99. International Joint ConferenceVolume 6, 10-16 July 1999 Page(s):4428 - 4431 vol.6 Digital Object Identifier 10.1109/IJCNN.1999.830883

【8】 Mantawy, A.H.; Abdel-Magid, Y.L.; Selim, S.Z.; " Integrating genetic algorithms, tabu search, and simulated annealing for the unit commitment problem" IEEE Transactions on Power Systems Volume 14, Issue 3, Aug. 1999 Page(s):829 – 836

【9】 H. Chen , A.F. Murray, "Continuous restricted Boltzmann machine with an implementable training algorithm" IEE proceedings-vision image and signal processing 150 (3): 153-158 JUN 2003
【10】G. E. Hinton, "Training products of experts by minimizing contrastive divergence". NeuralComputation, vol. 14, no. 8, pp. 1771.1800, 2002.

【11】J. Alspector, J.W. Gannett, S. Haber, M. B. Parker, and R. Chu, "A vlsi-ef_cient technique for generating multiple uncorrelated noise sources and its application to stochastic neural networks",. IEEE Trans. Circuits and Systems, vol. 38, no. 1, pp. 109.123, 1991.

【12】L. Chen and B. Shi, . "Cmos pwm vlsi implementation of neural network, ". in Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, vol. 3,pp. 485.488, July 2000.

【13】E. Vittoz, "Mos transistors operated in the lateral bipolar mode and their application in cmos technology". IEEE Journal of Solid-State Circuits, vol. SC-18, no. 3, pp. 273.279,1983.

【14】J. Lansner and T. Lehmann, "An analog cmos chip set for neural networks with arbitrary topologies". IEEE Transactions on Neural Networks, vol. 4, no. 3, pp. 441.444, 1993.

【15】M. Banu and Y. Tsividis, . "Floating voltage-controlled resistors in cmos technology. " Electronics Letters, vol. 18, pp. 678.679, 1982.

【16】J. Alspector, A. Jayakumar, and S. Luma, "Experimental evaluation of learning in a neural microsystem, "Advances in Neural Information Processing Systems (NIPS91), vol. 4,pp. 871.878, Dec 1992.

【17】J. Alspector, B. Gupta, and R. B. Allen, "Performance of a stochastic learning microchip",Advances in Neural Information Processing Systems(NIPS88), vol. 1, pp. 748.760,Dec 1989.

【18】Chible, H.; "Analog circuit for synapse neural networks VLSI implementation " The IEEE International Conference of Electronics Circuits and Systems，ICECS 2000 . Volume 2, 17-20 Dec. 2000 Page(s):1004 - 1007 vol.2

【19】 Hui-Chin Tseng; Chi-Sheng Lin; Hsin-Hung Ou; Bin-Da Liu;"A Low-Power Rail-to-Rail 6-bit Flash ADC Based on a Novel Complementary Average-Value Approach"Low Power Electronics and Design, 2004. ISLPED '04. Proceedings of the 2004 International Symposium on，2004 Page(s):252 – 256

【20】Steyaert, M.; Roovers, R.; Craninckx, J.; "A 100 MHz 8 bit CMOS interpolating A/D converter" Custom Integrated Circuits Conference, 1993., Proceedings of the IEEE 1993 9-12 May 1993 Page(s):28.1.1 - 28.1.4

【21】Roovers, R.; Steyaert, M.;Advanced"Design of CMOS A/D converters with folding and/or interpolating Techniques A-D and D-A Conversion Techniques and their Applications, "1994. Second International Conference on6-8 Jul 1994 Page(s):76 - 81

【22】Yin, G.M.; Eynde, F.O.; Sansen, W.; "A high-speed CMOS comparator with 8-b resolution"IEEE Journal of Solid-State Circuits Volume 27, Issue 2, Feb. 1992 Page(s):208 – 211

【23】Diotalevi, F.; Valle, M.; Bo, G.M.; Biglieri, E.; Caviglia, D.D. "Analog CMOS current mode neural primitives ";Circuits and Systems, 2000. Proceedings. ISCAS 2000 Geneva. The 2000 IEEE International Symposium on Volume 2, 28-31 May 2000 Page(s):717 - 720 vol.2

【24】Zinkevich, V.; "Analog memory devices based on NDRO toroidal cores" Magnetics, IEEE Transactions on Volume 10, Issue 1, Mar 1974 Page(s):89 - 92

【25】Urabe, T.; Sakurai, Y.; " Magnetic analog memory elements of a single core"Magnetics, IEEE Transactions on Volume 3, Issue 3, Sep 1967 Page(s):470 - 475