Neuromorphic Synapses for Artificial Dendrites Export

@article{citeulike:5224406, abstract = {We describe neuromorphic, variable-weight synapses onartificial dendrites that facilitate experimentation with correlativeadaptation rules. Attention is focused on those aspects of biologicalsynaptic function that could affect a neuromorphic network'scomputational power and adaptive capability. These include sublinearsummation, quantal synaptic noise, and independent adaptationof adjacent synapses. The diffusive nature of artificial dendritesadds considerable flexibility to the design of fast synapsesby allowing conductances to be enabled for short or for variablelengths of time. We present two complementary synapse designs:the shared conductance array and the self-timed synapse. Bothsynapse circuits behave as conductances to mimic biological synapsesand thus enable sublinear summation. The former achieves weightvariation by selecting different conductances from an on-chiparray, and the latter by modulating the length of time a constantconductance remains activated. Both work with our interchip communicationsystem, virtual wires. For the present purpose of comparing variousadaptation mechanisms in software, synaptic weights are storedoff chip. This simplifies the addition of quantal weight noiseand allows connections from different sources to the same dendriticcompartment to have independent weights.}, author = {Westerman, Wayne C. and Northmore, David P. M. and Elias, John}, citeulike-article-id = {5224406}, citeulike-linkout-0 = {http://dx.doi.org/10.1023/A:1008248431434}, citeulike-linkout-1 = {http://www.springerlink.com/content/r5n07j745457q07j}, day = {1}, doi = {10.1023/A:1008248431434}, journal = {Analog Integrated Circuits and Signal Processing}, keywords = {artificial, dendrite}, month = {May}, number = {1}, pages = {167--184}, posted-at = {2009-07-22 05:35:52}, priority = {2}, title = {Neuromorphic Synapses for Artificial Dendrites}, url = {http://dx.doi.org/10.1023/A:1008248431434}, volume = {13}, year = {1997} }

TY - JOUR ID - citeulike:5224406 L3 - citeulike-article-id:5224406 N2 - We describe neuromorphic, variable-weight synapses onartificial dendrites that facilitate experimentation with correlativeadaptation rules. Attention is focused on those aspects of biologicalsynaptic function that could affect a neuromorphic network'scomputational power and adaptive capability. These include sublinearsummation, quantal synaptic noise, and independent adaptationof adjacent synapses. The diffusive nature of artificial dendritesadds considerable flexibility to the design of fast synapsesby allowing conductances to be enabled for short or for variablelengths of time. We present two complementary synapse designs:the shared conductance array and the self-timed synapse. Bothsynapse circuits behave as conductances to mimic biological synapsesand thus enable sublinear summation. The former achieves weightvariation by selecting different conductances from an on-chiparray, and the latter by modulating the length of time a constantconductance remains activated. Both work with our interchip communicationsystem, virtual wires. For the present purpose of comparing variousadaptation mechanisms in software, synaptic weights are storedoff chip. This simplifies the addition of quantal weight noiseand allows connections from different sources to the same dendriticcompartment to have independent weights. IS - 1 JF - Analog Integrated Circuits and Signal Processing EP - 184 TI - Neuromorphic Synapses for Artificial Dendrites VL - 13 SP - 167 KW - artificial KW - dendrite AU - Westerman, Wayne AU - Northmore, David AU - Elias, John PY - 1997/05/01/ UR - http://dx.doi.org/10.1023/A:1008248431434 ER -