Fig. 1: Overview of conceptual circuit architecture.
From: The backpropagation algorithm implemented on spiking neuromorphic hardware
Feedforward activations of input (x), hidden (h) and output (o) layers are calculated by a feedforward module. Errors (e = t − o) are calculated from the output and the training signal (t). Errors are backpropagated through a feedback module with the same weights W2 for synapses between h and o, but in the opposite direction (mathematically expressed as the transpose, \({W}_{2}^{T}\)). Local gradients (d1, d2) are gated back into the feedforward circuit at appropriate times to accomplish potentiation or depression of appropriate weights.
