Abstract:Spiking neural networks (SNNs), inspired by the neuronal network in the brain, provide biologically relevant and low-power consuming models for information processing. Existing studies either mimic the learning mechanism of brain neural networks as closely as possible, for example, the temporally local learning rule of spike-timing-dependent plasticity (STDP), or apply the gradient descent rule to optimize a multilayer SNN with fixed structure. However, the learning rule used in the former is local and how the real brain might do the global-scale credit assignment is still not clear, which means that those shallow SNNs are robust but deep SNNs are difficult to be trained globally and could not work so well. For the latter, the nondifferentiable problem caused by the discrete spike trains leads to inaccuracy in gradient computing and difficulties in effective deep SNNs. Hence, a hybrid solution is interesting to combine shallow SNNs with an appropriate machine learning (ML) technique not requiring the gradient computing, which is able to provide both energy-saving and high-performance advantages. In this article, we propose a HybridSNN, a deep and strong SNN composed of multiple simple SNNs, in which data-driven greedy optimization is used to build powerful classifiers, avoiding the derivative problem in gradient descent. During the training process, the output features (spikes) of selected weak classifiers are fed back to the pool for the subsequent weak SNN training and selection. This guarantees HybridSNN not only represents the linear combination of simple SNNs, as what regular AdaBoost algorithm generates, but also contains neuron connection information, thus closely resembling the neural networks of a brain. HybridSNN has the benefits of both low power consumption in weak units and overall data-driven optimizing strength. The network structure in HybridSNN is learned from training samples, which is more flexible and effective compared with existing fixed multilayer SNNs. Moreover, the topological tree of HybridSNN resembles the neural system in the brain, where pyramidal neurons receive thousands of synaptic input signals through their dendrites. Experimental results show that the proposed HybridSNN is highly competitive among the state-of-the-art SNNs.

A smart hybrid memory scheduling approach using neural models

Intelligent Hybrid Memory Scheduling Based on Page Pattern Recognition

A Low-Cost and Pages-Interrelation-Aware Attention Model for Hybrid Memory Scheduling

HybridSNN: Combining Bio-Machine Strengths by Boosting Adaptive Spiking Neural Networks.

Stochastic Computation Based Hybrid Artificial Neural Networks and Spike Neural Network

A Hierarchical Neural Task Scheduling Algorithm in the Operating System of Neuromorphic Computers

Hierarchical Memory-Constrained Operator Scheduling of Neural Architecture Search Networks

Spatial-Temporal Hybrid Neural Network With Computing-in-Memory Architecture

Memory-aware Scheduling for Complex Wired Networks with Iterative Graph Optimization

AutoTM: Automatic Tensor Movement in Heterogeneous Memory Systems using Integer Linear Programming

Energy-Efficient Hybrid-RAM with Hybrid Bit-Serial Based VMM Support

Scaling Memory-Augmented Neural Networks with Sparse Reads and Writes

A Hybrid RRAM-SRAM Computing-In-Memory Architecture for Deep Neural Network Inference-Training Edge Acceleration

Towards Efficient Deployment of Hybrid SNNs on Neuromorphic and Edge AI Hardware

Spiking Hyperdimensional Network: Neuromorphic Models Integrated with Memory-Inspired Framework

Efficient Memory Management for Deep Neural Net Inference

Hybrid Neural Network Architecture for On-Line Learning

BARM: A Batch-Aware Resource Manager for Boosting Multiple Neural Networks Inference on GPUs With Memory Oversubscription

An Application-oblivious Memory Scheduling System for DNN Accelerators

Hybrid computing using a neural network with dynamic external memory

SmartMoE: Efficiently Training Sparsely-Activated Models through Combining Offline and Online Parallelization.