A More Accurate Approximation of Activation Function with Few Spikes Neurons
This addresses computational efficiency for deep neural networks by enhancing spiking neural networks, but it appears incremental as it builds on existing few spikes neurons.
The paper tackles the problem of approximating complex activation functions like Swish with energy-efficient spiking neurons, proposing a tendency-based parameter initialization method that improves accuracy, though no concrete numbers are provided.
Recent deep neural networks (DNNs), such as diffusion models [1], have faced high computational demands. Thus, spiking neural networks (SNNs) have attracted lots of attention as energy-efficient neural networks. However, conventional spiking neurons, such as leaky integrate-and-fire neurons, cannot accurately represent complex non-linear activation functions, such as Swish [2]. To approximate activation functions with spiking neurons, few spikes (FS) neurons were proposed [3], but the approximation performance was limited due to the lack of training methods considering the neurons. Thus, we propose tendency-based parameter initialization (TBPI) to enhance the approximation of activation function with FS neurons, exploiting temporal dependencies initializing the training parameters.