Photonic integrated neuro-synaptic core for convolutional spiking neural network

Neuromorphic photonic computing has emerged as a competitive computing paradigm to overcome the bottlenecks of the von-Neumann architecture.Linear weighting and nonlinear spike activation are two fundamental functions of a photonic spiking neural network(PSNN).However,they are separately implemented with different photonic materials and devices,hindering the large-scale integration of PSNN.Here,we propose,fabricate and experimentally demonstrate a photonic neuro-synaptic chip enabling the simultaneous implementation of linear weighting and nonlinear spike activation based on a distributed feedback(DFB)laser with a saturable absorber(DFB-SA).A prototypical system is experimentally constructed to demonstrate the parallel weighted function and nonlinear spike activation.Furthermore,a fourchannel DFB-SA laser array is fabricated for realizing matrix convolution of a spiking convolutional neural network,achieving a recognition accuracy of 87%for the MNIST dataset.The fabricated neuro-synaptic chip offers a fundamental building block to construct the large-scale integrated PSNN chip.

Multiple SiGe/Si layers epitaxy and SiGe selective etching for vertically stacked DRAM

Fifteen periods of Si/Si_(0.7)Ge_(0.3)multilayers(MLs)with various Si Ge thicknesses are grown on a 200 mm Si substrate using reduced pressure chemical vapor deposition(RPCVD).Several methods were utilized to characterize and analyze the ML structures.The high resolution transmission electron microscopy(HRTEM)results show that the ML structure with 20 nm Si_(0.7)Ge_(0.3)features the best crystal quality and no defects are observed.Stacked Si_(0.7)Ge_(0.3)ML structures etched by three different methods were carried out and compared,and the results show that they have different selectivities and morphologies.In this work,the fabrication process influences on Si/Si Ge MLs are studied and there are no significant effects on the Si layers,which are the channels in lateral gate all around field effect transistor(L-GAAFET)devices.For vertically-stacked dynamic random access memory(VS-DRAM),it is necessary to consider the dislocation caused by strain accumulation and stress release after the number of stacked layers exceeds the critical thickness.These results pave the way for the manufacture of high-performance multivertical-stacked Si nanowires,nanosheet L-GAAFETs,and DRAM devices.

Integration of α-fairness with DEA based resource allocation model

How to allocate a resource efficiently and fairly attracts the attention of both researchers and practitioners. Data envelopment analysis(DEA) has been brought to bear on its solution. The existing literature applies Gini coefficient to measure the fairness in the resource allocation process. However, the Gini coefficient is inapplicable in many applications. This paper proposes a novel centralized resource allocation model based on DEA that considers both the efficiency and the fairness. This paper adopts a notion of fairness, namely α-fairness that is well studied in welfare economics and is of practical significance. The new model integratesα-fairness with DEA to support resource allocation decisions. It aids decision makers in making a trade-off between the efficiency and the fairness. An illustrative application is used to validate the proposed approach.

Self-Supported Nanoporous Gold with Gradient Tin Oxide for Sustainable and Efficient Hydrogen Evolution in Neutral Media

Hydrogen evolution reaction(HER)in neutral medium suffers from slow kinetics as compared to that in alkaline or acidic conditions,owing to larger Ohmic loss and low proton concentration.Here we report that a self-supported nanoporous Au-SnO_(x)(NP Au-SnO_(x))catalyst with gradient tin oxide surface could significantly enhance HER activity in neutral buffer solution(0.2 M PBS).The NP Au-SnO_(x)catalyst exhibits a low onset overpotential of 38 mV and a small Tafel slope of 79 mV dec^(−1).The current density of 10 mA cm^(−2)is manifested at an overpotential as low as 148 mV,representing the comparable performance of Pt/C catalyst.This high catalytic activity can retain at least 10 hours without any detectable decay.The superior HER activity is proposed to originate from the gradient SnO_(x)structure and metal/oxide interfaces in nanoporous ligaments.Furthermore,the X-ray photoelectron spectroscopy reveals that the gradient oxide in the ligament is remarkably stable during long-term reaction.

Variant rs2237892 of KCNQ1 Is Potentially Associated with Hypertension and Macrovascular Complications in Type 2 Diabetes Mellitus in A Chinese Han Population

KCNQ1 has been identified as a susceptibility gene of type 2 diabetes mellitus(T2DM) in Asian populations through genome-wide association studies. However, studies on the association between gene polymorphism of KCNQ1 and T2 DM complications remain unclear. To further analyze the association between different alleles at the single nucleotide polymorphism(SNP) rs2237892 within KCNQ1 and TD2 M and its complications, we conducted a case-control study in a Chinese Han population. The C allele of rs2237892 variant contributed to susceptibility to T2DM(odds ratio [OR], 1.45; 95% confidence interval [CI], 1.20–1.75). Genotypes CT(OR, 1.97; 95% CI,1.24–3.15) and CC(OR, 2.49; 95% CI, 1.57–3.95) were associated with an increased risk of T2 DM. Multivariate regression analysis was performed with adjustment of age, gender, and body mass index. We found that systolic blood pressure(P = 0.015), prevalence of hypertension(P = 0.037), and risk of macrovascular disease(OR, 2.10; CI, 1.00–4.45) were significantly higher in subjects with the CC genotype than in the combined population with genotype either CT

Nanoporous Au-Sn with solute strain for simultaneously enhanced selectivity and durability during electrochemical CO2 reduction

Electrochemical carbon dioxide reduction meditated by metallic catalysts suffers from restricted selectivity and competition from hydrogen evolution, which sensitively depends on ambiguous contributions of alloying and strain state in bimetallic catalysts. Herein, nanoporous Au-Sn(NPAS) containing trace tin solute in Au lattices is delicately designed to convince real strain effect, while eliminating other undesirable factors, such as alloying, crystal facets and surface composition. Compared with nanoporous gold(NPG), the NPAS with a solute strain of ~2.2% enables more efficient CO2-to-CO conversion, with an efficiency as high as 92% at-0.85 V versus reversible hydrogen electrode(vs. RHE), and the high activity can retain for more than 8 h. The combination of HRTEM and surface valence band photoemission spectra reveals that the tensile strain on the surface of 3 D nanoporous structure promotes the catalytic activity by shifting up the d-band center and strengthening the adsorption of key intermediate *COOH. A small amount of Sn solute in the nanoporous alloy can prevent ligament coarsening effectively and improve the electrochemical stability.