Deep Spiking Networks
This work addresses the problem of efficient learning on streaming event-based data for researchers in neuromorphic computing, though it is incremental as it builds on existing spiking network concepts.
The authors tackled the challenge of training deep spiking neural networks by introducing a backpropagation algorithm that makes them behave identically to conventional deep networks with rectified-linear units in the limit of long run times, achieving performance very close to a conventional Multi-Layer Perceptron on MNIST classification.
We introduce an algorithm to do backpropagation on a spiking network. Our network is "spiking" in the sense that our neurons accumulate their activation into a potential over time, and only send out a signal (a "spike") when this potential crosses a threshold and the neuron is reset. Neurons only update their states when receiving signals from other neurons. Total computation of the network thus scales with the number of spikes caused by an input rather than network size. We show that the spiking Multi-Layer Perceptron behaves identically, during both prediction and training, to a conventional deep network of rectified-linear units, in the limiting case where we run the spiking network for a long time. We apply this architecture to a conventional classification problem (MNIST) and achieve performance very close to that of a conventional Multi-Layer Perceptron with the same architecture. Our network is a natural architecture for learning based on streaming event-based data, and is a stepping stone towards using spiking neural networks to learn efficiently on streaming data.