🤖 AI Summary
To address performance bottlenecks in Spiking Neural Networks (SNNs) when processing complex temporal information—stemming from rigid neuron models and limited encoding schemes—this paper proposes: (1) an Unconstrained Leaky Integrate-and-Fire (ULIF) neuron model, which introduces learnable, time-varying membrane potential parameters to enhance multi-scale temporal dynamics modeling; and (2) a hybrid spatiotemporal encoding mechanism that jointly leverages rate and temporal coding, explicitly capturing temporal structure while improving information representation efficiency. The proposed framework supports end-to-end differentiable training. Experimental results demonstrate significant improvements in both accuracy and energy efficiency on benchmark object detection and recognition tasks. To foster reproducibility and further research, the implementation code is publicly released.
📝 Abstract
Spiking Neural Networks (SNN) exhibit higher energy efficiency compared to Artificial Neural Networks (ANN) due to their unique spike-driven mechanism. Additionally, SNN possess a crucial characteristic, namely the ability to process spatio-temporal information. However, this ability is constrained by both internal and external factors in practical applications, thereby affecting the performance of SNN. Firstly, the internal issue of SNN lies in the inherent limitations of their network structure and neuronal model, which result in the network adopting a unified processing approach for information of different temporal dimensions when processing input data containing complex temporal information. Secondly, the external issue of SNN stems from the direct encoding method commonly adopted by directly trained SNN, which uses the same feature map for input at each time step, failing to fully exploit the spatio-temporal characteristics of SNN. To address these issues, this paper proposes an Unconstrained Leaky Integrate-and-Fire (ULIF) neuronal model that allows for learning different membrane potential parameters at different time steps, thereby enhancing SNN' ability to process information of different temporal dimensions. Additionally, this paper presents a hybrid encoding scheme aimed at solving the problem of direct encoding lacking temporal dimension information. Experimental results demonstrate that the proposed methods effectively improve the overall performance of SNN in object detection and object recognition tasks. related code is available at https://github.com/hhx0320/ASNN.