Abstract:Spiking neural networks (SNNs) based on the leaky integrate and fire (LIF) model have been applied to energy-efficient temporal and spatiotemporal processing tasks. Due to the bioplausible neuronal dynamics and simplicity, LIF-SNN benefits from event-driven processing, however, usually face the embarrassment of reduced performance. This may because, in LIF-SNN, the neurons transmit information via spikes. To address this issue, in this work, we propose a leaky integrate and analog fire (LIAF) neuron model so that analog values can be transmitted among neurons, and a deep network termed LIAF-Net is built on it for efficient spatiotemporal processing. In the temporal domain, LIAF follows the traditional LIF dynamics to maintain its temporal processing capability. In the spatial domain, LIAF is able to integrate spatial information through convolutional integration or fully connected integration. As a spatiotemporal layer, LIAF can also be used with traditional artificial neural network (ANN) layers jointly. In addition, the built network can be trained with backpropagation through time (BPTT) directly, which avoids the performance loss caused by ANN to SNN conversion. Experiment results indicate that LIAF-Net achieves comparable performance to the gated recurrent unit (GRU) and long short-term memory (LSTM) on bAbI question answering (QA) tasks and achieves state-of-the-art performance on spatiotemporal dynamic vision sensor (DVS) data sets, including MNIST-DVS, CIFAR10-DVS, and DVS128 Gesture, with much less number of synaptic weights and computational overhead compared with traditional networks built by LSTM, GRU, convolutional LSTM (ConvLSTM), or 3-D convolution (Conv3D). Compared with traditional LIF-SNN, LIAF-Net also shows dramatic accuracy gain on all these experiments. In conclusion, LIAF-Net provides a framework combining the advantages of both ANNs and SNNs for lightweight and efficient spatiotemporal information processing.

GLIF: A Unified Gated Leaky Integrate-and-Fire Neuron for Spiking Neural Networks

IM-LIF: Improved Neuronal Dynamics with Attention Mechanism for Direct Training Deep Spiking Neural Network

Adaptive Deep Spiking Neural Network with Global-Local Learning Via Balanced Excitatory and Inhibitory Mechanism

CLIF: Complementary Leaky Integrate-and-Fire Neuron for Spiking Neural Networks

An Efficient Learning Algorithm for Direct Training Deep Spiking Neural Networks

Deep Spiking Neural Network Using Spatio-temporal Backpropagation with Variable Resistance.

Sharing Leaky-Integrate-and-Fire Neurons for Memory-Efficient Spiking Neural Networks

A Brain-Inspired Spiking Neural Network Model with Temporal Encoding and Learning

Learnable Surrogate Gradient for Direct Training Spiking Neural Networks

KLIF: An Optimized Spiking Neuron Unit for Tuning Surrogate Gradient Function

LIAF-Net: Leaky Integrate and Analog Fire Network for Lightweight and Efficient Spatiotemporal Information Processing

Adaptive Multi-Level Firing for Direct Training Deep Spiking Neural Networks

Reconstruction of Adaptive Leaky Integrate-and-Fire Neuron to Enhance the Spiking Neural Networks Performance by Establishing Complex Dynamics

Learning High-Performance Spiking Neural Networks with Multi-Compartment Spiking Neurons.

Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning

Unleashing the Potential of Spiking Neural Networks for Sequential Modeling with Contextual Embedding.

Memristive leaky integrate-and-fire neuron and learnable straight-through estimator in spiking neural networks

Incorporating Learnable Membrane Time Constant to Enhance Learning of Spiking Neural Networks

Hierarchical Spiking-Based Model for Efficient Image Classification with Enhanced Feature Extraction and Encoding.

SGLFormer: Spiking Global-Local-Fusion Transformer with high performance

Brain-inspired Multilayer Perceptron with Spiking Neurons