Spiking Approximations of the MaxPooling Operation in Deep SNNs
This work addresses a specific bottleneck for researchers and engineers aiming to convert CNNs to low-power SNNs, though it is incremental as it builds on existing ANN-to-SNN conversion methods.
The paper tackles the lack of hardware-friendly spiking equivalents for MaxPooling layers in deep Spiking Neural Networks (SNNs), which limits ANN-to-SNN conversion, by presenting two methods to implement MaxPooling in SNNs and demonstrating their feasibility on Intel's Loihi neuromorphic hardware with datasets like MNIST, FMNIST, and CIFAR10.
Spiking Neural Networks (SNNs) are an emerging domain of biologically inspired neural networks that have shown promise for low-power AI. A number of methods exist for building deep SNNs, with Artificial Neural Network (ANN)-to-SNN conversion being highly successful. MaxPooling layers in Convolutional Neural Networks (CNNs) are an integral component to downsample the intermediate feature maps and introduce translational invariance, but the absence of their hardware-friendly spiking equivalents limits such CNNs' conversion to deep SNNs. In this paper, we present two hardware-friendly methods to implement Max-Pooling in deep SNNs, thus facilitating easy conversion of CNNs with MaxPooling layers to SNNs. In a first, we also execute SNNs with spiking-MaxPooling layers on Intel's Loihi neuromorphic hardware (with MNIST, FMNIST, & CIFAR10 dataset); thus, showing the feasibility of our approach.