On lower iteration complexity bounds for the convex concave saddle point problems

Mathematical Programming - In this paper, we study the lower iteration complexity bounds for finding the saddle point of a strongly convex and strongly concave saddle point problem: $$\min _x\max...     

基于GA_XGBoost模型的大学生科研能力预测问题研究

科学评价大学生科研创新能力对我国科研水平的提高具有重要意义.采用机器学习模型来预测大学生科研能力可以起到良好的效果,提出一种GAXGBoost模型来实现对大学生的科研能力预测.此模型是以Xgboost算法为基础,然后充分利用遗传算法的全局搜索能力自动搜索Xgboost最优超参数,避免了人为经验调参不准确的缺陷,最后采用精英选择策略以此确保每一轮都是最佳的进化结果.通过分析表明,所采用的GAXGBoost模型在大学生科研能力预测的结果中具有很高的精度,将此模型与Logistic Regression、Random Forest、SVM等模型进行对比,GAXGBoost模型的预测精度最高.     

Fabricating Dual-Atom Iron Catalysts for Efficient Oxygen Evolution Reaction: A Heteroatom Modulator Approach

Understanding the thermal aggregation behavior of metal atoms is important for the synthesis of supported metal clusters. Here, derived from a metal–organic framework encapsulating a trinuclear Fe^III₂Fe^II complex (denoted as Fe₃) within the channels, a well-defined nitrogen-doped carbon layer is fabricated as an ideal support for stabilizing the generated iron nanoclusters. Atomic replacement of Fe^II by other metal(II) ions (e.g., Zn^II/Co^II) via synthesizing isostructural trinuclear-complex precursors (Fe₂Zn/Fe₂Co), namely the “heteroatom modulator approach”, is inhibiting the aggregation of Fe atoms toward nanoclusters with formation of a stable iron dimer in an optimal metal–nitrogen moiety, clearly identified by direct transmission electron microscopy and X-ray absorption fine structure analysis. The supported iron dimer, serving as cooperative metal–metal site, acts as efficient oxygen evolution catalyst. Our findings offer an atomic insight to guide the future design of ultrasmall metal clusters bearing outstanding catalytic capabilities.     

Boron(iii) β-diketonate-based small molecules for functional non-fullerene polymer solar cells and organic resistive memory devices

A class of acceptor–donor–acceptor chromophoric small-molecule non-fullerene acceptors, 1–4, with difluoroboron(iii) β-diketonate (BF₂bdk) as the electron-accepting moiety has been developed. Through the variation of the central donor unit and the modification on the peripheral substituents of the terminal BF₂bdk acceptor unit, their photophysical and electrochemical properties have been systematically studied. Taking advantage of their low-lying lowest unoccupied molecular orbital energy levels (from −3.65 to −3.72 eV) and relatively high electron mobility (7.49 × 10⁻⁴ cm² V⁻¹ s⁻¹), these BF₂bdk-based compounds have been employed as non-fullerene acceptors in organic solar cells with maximum power conversion efficiencies of up to 4.31%. Moreover, bistable resistive memory characteristics with charge-trapping mechanisms have been demonstrated in these BF₂bdk-based compounds. This work not only demonstrates for the first time the use of a boron(iii) β-diketonate unit in constructing non-fullerene acceptors, but also provides more insights into designing organic materials with multi-functional properties.

Boron(iii) β-diketonates have been demonstrated to serve as multi-functional materials in NFA-based OPVs and organic resistive memories.     

Close‐Packed Two‐Dimensional Silver Nanoparticle Arrays: Quadrupolar and Dipolar Surface Plasmon Resonance Coupling

Silver nanoparticles (NPs) ranging in size from 40 to 100 nm were prepared in high yield by using an improved seed‐mediated method. The homogeneous Ag NPs were used as building blocks for 2D assembled Ag NP arrays by using an oil/water interface. A close‐packed 2D array of Ag NPs was fabricated by using packing molecules (3‐mercaptopropyltrimethoxysilane) to control the interparticle spacing. The homogeneous 2D Ag NP array exhibited a strong quadrupolar cooperative plasmon mode resonance and a dipolar red‐shift relative to individual Ag NPs suspended in solution. A well‐arranged 2D Ag NP array was embedded in polydimethylsiloxane film and, with biaxial stretching to control the interparticle distance, concomitant variations of the quadrupolar and dipolar couplings were observed. As the interparticle distance increased, the intensity of the quadrupolar cooperative plasmon mode resonance decreased and dipolar coupling completely disappeared. The local electric field of the 2D Ag NP array was calculated by using finite difference time domain simulation and qualitatively showed agreement with the experimental measurements.     

Feasibility study on buoyancy–weight ratios of a submerged floating tunnel prototype subjected to hydrodynamic loads

The research progress of a novel traffic solution,a submerged floating tunnel(SFT),is reviewed in terms of a study approach and loading scenario.Among existing publications,the buoyancy-weight ratio(BWR) is usually predefined.However,BWR is a critical structural parameter that tremendously affects the dynamic behaviour of not only the tunnel tube itself but also the cable system.In the context of a SFT prototype(SFTP) project in Qiandao Lake(Zhejiang Province,China),the importance of BWR is illustrated by finite element analysis and subsequently,an optimized BWR is proposed within a reasonable range in the present study.In the numerical model,structural damping is identified to be of importance.Rayleigh damping and the corresponding Rayleigh coefficients are attained through a sensitivity study,which shows that the adopted damping ratios are fairly suitable for SFTP.Lastly,the human sense of security is considered by quantifying the comfort index,which helps further optimize BWR in the SFTP structural parameter design.     

紫色球杆菌视紫红质光谱特性的机器学习研究

近年来，机器学习等人工智能技术被应用于蛋白质工程，其在蛋白质结构、功能预测、催化活性等研究中具有独特优势。在未知蛋白质结构的情况下，将蛋白质序列和功能特性与机器学习相结合，基于序列-活性关系（innovative sequence-activity relationship，ISAR）算法，将蛋白质氨基酸序列数字化，用快速傅里叶变换（fast four transform，FFT）进行预处理，再进行偏最小二乘回归建模，可在数据集较少情况下拟合得到最佳模型。通过机器学习对紫色球杆菌视紫红质（gloeobacter violaceus rhodopsin，GR）的突变体蛋白质氨基酸序列与光谱最大吸收波长进行建模，获得了最佳模型。用最佳索引LEVM760106建模得到的确定系数R² 为0.944，均方误差E为11.64。用小波变换进行的预处理，其R² 虽也约为0.944，但E大于11.64，不及FFT进行的预处理。方法较好地解决了蛋白质序列与功能特性之间的数学建模问题，在蛋白质工程中可为预测更优的突变体提供支持。