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

Introduction: Spiking Neural Networks (SNNs), inspired by brain science, offer low energy consumption and high biological plausibility with their event-driven nature. However, the current SNNs are still suffering from insufficient performance. Methods: Recognizing the brain's adeptness at information processing for various scenarios with complex neuronal connections within and across regions, as well as specialized neuronal architectures for specific functions, we propose a Spiking Global-Local-Fusion Transformer (SGLFormer), that significantly improves the performance of SNNs. This novel architecture enables efficient information processing on both global and local scales, by integrating transformer and convolution structures in SNNs. In addition, we uncover the problem of inaccurate gradient backpropagation caused by Maxpooling in SNNs and address it by developing a new Maxpooling module. Furthermore, we adopt spatio-temporal block (STB) in the classification head instead of global average pooling, facilitating the aggregation of spatial and temporal features. Results: SGLFormer demonstrates its superior performance on static datasets such as CIFAR10/CIFAR100, and ImageNet, as well as dynamic vision sensor (DVS) datasets including CIFAR10-DVS and DVS128-Gesture. Notably, on ImageNet, SGLFormer achieves a top-1 accuracy of 83.73% with 64 M parameters, outperforming the current SOTA directly trained SNNs by a margin of 6.66%. Discussion: With its high performance, SGLFormer can support more computer vision tasks in the future. The codes for this study can be found in https://github.com/ZhangHanN1/SGLFormer.

Direct training high-performance deep spiking neural networks: a review of theories and methods.

Spiking neural networks (SNNs) offer a promising energy-efficient alternative to artificial neural networks (ANNs), in virtue of their high biological plausibility, rich spatial-temporal dynamics, and event-driven computation. The direct training algorithms based on the surrogate gradient method provide sufficient flexibility to design novel SNN architectures and explore the spatial-temporal dynamics of SNNs. According to previous studies, the performance of models is highly dependent on their sizes. Recently, direct training deep SNNs have achieved great progress on both neuromorphic datasets and large-scale static datasets. Notably, transformer-based SNNs show comparable performance with their ANN counterparts. In this paper, we provide a new perspective to summarize the theories and methods for training deep SNNs with high performance in a systematic and comprehensive way, including theory fundamentals, spiking neuron models, advanced SNN models and residual architectures, software frameworks and neuromorphic hardware, applications, and future trends.

High electrochemical performance based on ultrathin porous CuO nanobelts grown on Cu substrate as integrated electrode.

A facile and low-cost approach has been developed to fabricate porous CuO nanobelts directly grown on a Cu substrate. The as-prepared CuO samples can be directly used as integrated electrodes for lithium-ion batteries and pseudo-supercapacitors without the addition of other ancillary materials such as carbon black or a binder to enhance electrode conductivity and cycling stability. The unique nanostructural features endow them with excellent electrochemical performance as demonstrated by high capacities of 640 mA h g(-1) after 100 cycles at 0.2 C rate and an excellent specific capacitance of 340 F g(-1), which corresponds to the energy density of 45 W h kg(-1). The cyclability of the electrode demonstrates only a 10-15% loss in capacitance over 5000 cycles.

Communication speeds up but impairs the consensus decision in a dyadic colour estimation task.

Communication plays an important role in consensus decision-making which pervades our daily life. However, the exact role of communication in consensus formation is not clear. Here, to study the effects of communication on consensus formation, we designed a dyadic colour estimation task, where a pair of isolated participants repeatedly estimated the colours of discs until they reached a consensus or completed eight estimations, either with or without communication. We show that participants' estimates gradually approach each other, reaching towards a consensus, and these are enhanced with communication. We also show that dyadic consensus estimation is on average better than individual estimation. Surprisingly, consensus estimation without communication generally outperforms that with communication, indicating that communication impairs the improvement of consensus estimation. However, without communication, it takes longer to reach a consensus. Moreover, participants who partially cooperate with each other tend to result in better overall consensus. Taken together, we have identified the effect of communication on the dynamics of consensus formation, and the results may have implications on group decision-making in general.

Flexible superior electrode architectures based on three-dimensional porous spinous α-Fe2O3 with a high performance as a supercapacitor.

Flexible supercapacitors have recently attracted increasing attention as they show unique promising advantages, such as flexibility and shape diversity, and they are light-weight and so on. Herein, we designed a series of 3D porous spinous iron oxide materials synthesized on a thin iron plate through a facile method under mild conditions. The unique nanostructural features endow them with excellent electrochemical performance. The electrochemical properties of the integrated electrodes as active electrode materials for supercapacitors have been investigated using different electrochemical techniques including cyclic voltammetry, and galvanostatic charge-discharge in Na2SO4 and LiPF6/EC : DEC electrolyte solutions. These integrated electrodes showed high specific capacitance (as high as 524.6 F g(-1) at the current density of 1 A g(-1)) in 1.0 M Na2SO4 (see Table S1). Moreover, the integrated electrodes also show high power densities and high energy densities in a LiPF6/EC : DEC electrolyte solution; for example, the energy densities were 319.3, 252.5, 152.1, 74.13 and 38.6 W h kg(-1) at different power densities of 8.81, 21.59, 56.65, 92.09 and 152.64 kW kg(-1), respectively. Additionally, the flexible superior electrode exhibited excellent stability with capacitance retention of 92.9% after 5000 cycles. Therefore, such flexible integrated devices might be used in smart and portable electronics.

Effective electrocatalysis based on Ag2O nanowire arrays supported on a copper substrate.

Silver oxide nanowire arrays (Ag2O NWAs) were first synthesized on a copper (Cu) rod by a simple and facile wet-chemistry approach without using any surfactants. The as-synthesized Ag2O NWA/Cu rod not only can be used as an integrated electrode (called a Ag2O NWA/CRIE) to detect hydrazine (HZ) but also can serve as the catalyst layer for a direct HZ fuel cell. The current density of HZ oxidation on Ag2O NWA (94.4 mA cm(-2)) is much bigger than that on a bare Cu rod (3.9 mA cm(-2)) at -0.6 V, and other Ag2O NWAs have the lowest onset potential (-0.85 V). This suggests that a Ag2O NWA integrated electrode has potential application in catalytic fields that contain the HZ fuel cell.