Analog implementation for networks of integrate-and-fire neurons with adaptive local connectivity

J. Schreiter1, U. Ramacher2, A. Heittmann2, D. Matolini, R. Schuffny1
1Department of Electrical Engineering and Information Technology, Dresden University of Technology, Dresden, Germany
2Corporate Research, Infineon AG, Munich, Germany

Tóm tắt

An analog VLSI implementation for pulse coupled neural networks of leakage free integrate-and-fire neurons with adaptive connections is presented. Weight adaptation is based on existing adaptation rules for image segmentation. Although both integrate-and-fire neurons and adaptive weights can be implementation only approximately, simulations have shown, that synchronization properties of the original adaptation rules are preserved.

Từ khóa

#Neurons #Image segmentation #Frequency synchronization #Very large scale integration #Hardware #Neural networks #Robustness #Signal processing #Nearest neighbor searches #Information technology

Tài liệu tham khảo

10.1162/neco.1996.8.2.373 10.1109/72.761708 varel, 1997, A quantitative description of short-term plasticity at exitatory synapses in layer 2/3 of rat-primary visual cortex, Neuroscience, 7926, 10.1523/JNEUROSCI.17-20-07926.1997 vittoz, 1997, Handbook of Neural Computation zador, 1998, Pulsed Neural Networks, Computing and Learning with Dynamic Synapses, 321 delbrueck, 1993, Bump Circuits, Cns memo 26 10.1109/MN.1999.758893 enz, 1995, MOS Modeling for LV-LI Circuit Design. CMOS & BICMOS IC Design, Techn report edwards, 1994, VLSI Phase Locking Architectures for Feature Linking in Multiple Target Tracking Systems, Advances in neural information processing systems, 6 henkel, 1998, A Simple and Fast Neural Network Approach to Stereovision, Proc of the Conf on Neural Information Processing Systems NIPS'97, 808 10.1007/3-540-46084-5_209 10.1109/72.870043 10.1007/BF00166411 10.1073/pnas.250483697