O'Reilly logo

Event-Based Neuromorphic Systems by Rodney Douglas, Adrian Whatley, Giacomo Indiveri, Tobi Delbruck, Shih-Chii Liu

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

8

Silicon Synapses

images

Synapses form the connections between biological neurons and so can form connections between the silicon neurons described in Chapter 7. They are fundamental elements for computation and information transfer in both real and artificial neural systems. While modeling, the nonlinear properties and the dynamics of real synapses can be extremely onerous for software simulations, neuromorphic very large scale integration (VLSI) circuits efficiently reproduce synaptic dynamics in real-time. VLSI synapse circuits convert input spikes into analog charge which then produces post-synaptic currents that get integrated at the membrane capacitance of the post-synaptic neuron. This chapter discusses various circuit strategies used in implementing the temporal dynamics observed in real synapses. It also presents circuits that implement nonlinear effects with short-term dynamics, such as short-term depression and facilitation, as well as long-term dynamics such as spike-based learning mechanisms for updating the weight of a synapse for learning systems such as the ones discussed in Chapter 6 and the nonvolatile weight storage and update using floating-gate technology discussed in Chapter 10.

__________

The picture is an electron micrograph of mouse somatosensory cortex, arrowheads point to post-synaptic densities. Reproduced with permission of Nuno da Costa, John Anderson, and ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required