Delay Neural Networks (DeNN) for exploiting temporal information in event-based datasets
This work addresses the challenge of efficiently processing temporal data in neural networks, particularly for event-based audio and visual applications, though it appears incremental as it builds on existing neural network paradigms.
The authors tackled the problem of exploiting temporal information in event-based datasets by introducing Delay Neural Networks (DeNN), which compute neuron information based on synaptic delays and spike times, resulting in good performance with fewer parameters and less energy than other models.
In Deep Neural Networks (DNN) and Spiking Neural Networks (SNN), the information of a neuron is computed based on the sum of the amplitudes (weights) of the electrical potentials received in input from other neurons. We propose here a new class of neural networks, namely Delay Neural Networks (DeNN), where the information of a neuron is computed based on the sum of its input synaptic delays and on the spike times of the electrical potentials received from other neurons. This way, DeNN are designed to explicitly use exact continuous temporal information of spikes in both forward and backward passes, without approximation. (Deep) DeNN are applied here to images and event-based (audio and visual) data sets. Good performances are obtained, especially for datasets where temporal information is important, with much less parameters and less energy than other models.