Adaptive structure generation and neuronal differentiation for memory encoding in SNNs.

Memory is the core of cognition. The exploration of the memory encoding mechanism or the representation mechanism of information in the Spiking Neural Network (SNN) is the basis for the in-depth study of memory. In this paper, we study the memory encoding mechanism of multilayer SNN models from a biomimetic perspective and explore a method using the high biological likelihood of SNN to enable the network to effectively simulate memory effects. We proposed a series of heuristic neuron-growing connection algorithms and supervised network weight learning algorithms, which were applied to the unsupervised and supervised training process of the presentation layer. These methods optimized the structure of the representation layer, achieved functional differentiation of neurons, and enabled the network to generate differentiated representations for different data modes. Under our algorithm, the proposed model achieves stable convergence with identical pattern inputs, demonstrating distinct representations and sensitivities to different visual modalities. To achieve stable information expression within the network, we conducted various comparative experiments to determine diverse parameters of the complex network. This paper contributes to the development of Brain-inspired Intelligence by bridging the gap between computer science and neuroscience by using simulations to validate biological hypotheses and guide machine learning. [Display omitted] • This study proposes and constructs a multi-layer spiking neural network model with input, representation, supervision, and observation layers, in which the structure enables our model to exhibit high information capacity and the ability to simulate biological nervous systems, well connecting computer science and biological neuroscience. • This study uses a series of unsupervised connection generating algorithms and bipolar supervised learning algorithm to optimize the structure of the representation layer, achieve functional differentiation of neurons, and enable the network to generate differentiated representations for different data modes. • This study analyzes the abnormal states that are prone to occur in complex spiking neural networks and proposes solutions for situations where the network enters a paralyzed state. The relevant methods provide a new technical path for improving the robustness and stability of spiking neural networks. [ABSTRACT FROM AUTHOR]