基于门控联合池化自编码器的通用性文本表征

为了学习文本的语义表征,以往的研究者主要依赖于复杂的循环神经网络(recurrent neural networks, RNNs)和监督式学习方法。该文提出了一种门控联合池化自编码器(gated mean-max AAE)用于学习中英文的文本语义表征。该文的自编码器完全通过多头自注意力机制(multi-head self-attention mechanism)来构建编码器和解码器网络。在编码阶段,提出了均值—最大化(mean-max)联合表征策略,即同时运用平均池化(mean pooling)和最大池化(max pooling)操作来捕获输入文本中多样性的语义信息。为促使联合池化表征可以全面地指导重构过程,解码器采用门控操作进行动态关注。通过在大规模中英文未标注语料上训练模型,获得了高质量的句子编码器。在重构文本段落的实验中,该文模型在实验效果和计算效率上均超越了传统的RNNs模型。将公开训练好的文本编码器,使其可以方便地运用于后续的研究。     

从中西色彩观差异看中西方设计色彩的异与同

目的对中西方设计色彩差异进行理论分析。方法以中西色彩表达的文化差异为基础,通过对中西方色彩观的文化渊源进行比较分析,提炼出对中西传统色彩观的文化特点概括。结论通过对色彩在中西方建筑、服饰、装饰等设计中的运用案例进行分析,把握传统色彩观在历史空间中的演进与中西融合趋势,以求对21世纪全球语境下的现代设计色彩发展,提供基于中西传统色彩观特点比较分析基础上的把握。     

中国观念艺术发展症候分析

中国观念艺术三十多年的发展,是一场"没有坐标的运动。"中国观念艺术曾在一定时期和范围内偏离了人本主义思想的轨迹,没有沿着健康的道路发展。本文将从"后殖民化"与"本土化""智力"与"智慧""大众"与"精英"三悖论,对中国观念艺术发展的症候进行分析。     

Influence of LiBO2 addition on the microstructure and lithium-ion conductivity of Li1+xAlxTi2−x(PO4)3(x = 0.3) ceramic electrolyte

Concerning the safety problems of conventional Li-ion batteries with liquid electrolytes, it is crucial to develop reliable solid-state electrolytes with high ionic conductivity. Li_1+xAl_xTi_2?x(PO₄)₃ (LATP, x = 0.3) is regarded as one of the most promising solid electrolytes due to its high ionic conductivity and excellent chemical stability to humidity.Herein, a new strategy is proposed for improving the sintering behavior and enhancing the ionic conductivity of LATP by using LiBO₂ as the sintering aid via liquid phase sintering. The as-prepared sample LATP with homogeneous microstructure and high relative density of 97.1% was successfully synthesized, yielding high total ionic conductivity of 3.5 × 10^?4 S cm^?1 and low activation energy of 0.39 eV at room temperature. It was found that the addition of LiBO₂ could effectively enhance the densification and increase the ionic conductivity of LATP electrolyte, proving an effective way to synthesis LATP ceramics by a simple and reliable route.     

基于深度学习的安全缺陷报告预测方法实证研究

软件安全问题的发生在大多数情况下会造成非常严重的后果，及早发现安全问题，是预防安全事故的关键手段之一.安全缺陷报告预测可以辅助开发人员及早发现被测软件中潜藏的安全缺陷，从而尽早得以修复.然而，由于安全缺陷在实际项目中的数量较少，而且特征复杂（即安全缺陷类型繁多，不同类型安全缺陷特征差异性较大），这使得手工提取特征相对困难，并随后造成传统机器学习分类算法在安全缺陷报告预测性能方面存在一定的瓶颈.针对该问题，提出基于深度学习的安全缺陷报告预测方法，采用深度文本挖掘模型TextCNN和TextRNN构建安全缺陷报告预测模型；针对安全缺陷报告文本特征，使用skip-grams方式构建词嵌入矩阵，并借助注意力机制对TextRNN模型进行优化.所构建的模型在5个不同规模的安全缺陷报告数据集上展开了大规模实证研究，实证结果表明：深度学习模型在80%的实验案例中都要优于传统机器学习分类算法，性能指标F1-score平均可提升0.258，在最好的情况下甚至可以提升0.535.除此之外，针对安全缺陷报告数据集存在的类不均衡问题，对不同采样方法进行了实证研究，并对结果进行了分析.     

Indoor solid fuel use for heating and cooking with blood pressure and hypertension: A cross-sectional study among middle-aged and older adults in China

A cross-sectional study was conducted to investigate the impact of solid fuel use for heating and cooking on blood pressure (BP) and hypertension, using data from the China Health and Retirement Longitudinal Study (CHARLS). The primary fuels used for indoor heating and cooking were collected by questionnaires, respectively. Hypertension was defined based on self-report of physician's diagnosis, and/or measured BP, and/or anti-hypertensive medication use. Multivariate logistic regression models were constructed to assess the associations. Among 10 450 eligible participants, 68.2% and 57.2% used indoor solid fuel for heating and cooking, respectively. Compared with none/clean fuel users, solid fuel for heating was associated with elevated BP (adjusted β: 2.02, 95% CI: 1.04–3.01 for systolic BP; adjusted β: 1.36, 95% CI: 0.78–1.94 for diastolic BP) and increased risk of hypertension (adjusted odds ratio: 1.15, 95% CI: 1.03–1.29). The impact of indoor solid fuel for heating on BP was more evident in rural and north residents, and hypertensive patients. We did not detect any significant associations between solid fuel use for cooking and BP/hypertension. Indoor solid fuel use is prevalent in China, especially in the rural areas. Its negative impact on BP suggested that modernization of household fuel use may help to reduce the burden of hypertension in China.     

Patterns and stability of food preferences among a national representative sample of young,middle-aged,and elderly adults in China: A latent transition analysis

The purpose of this study was to explore the patterns and stability of, as well as the predictors for, Chinese adults’ food preferences. With the panel data set derived from the China Health and Nutrition Survey (CHNS), latent class analysis (LCA) and latent transition analysis (LTA) were employed to analyze the data of 8850 adults aged 18–40 years (young), 41–60 years (middle-aged) and 61 years above (elderly) over a 4-year period (year 2011 as baseline and year 2015 as follow-up). Latent class analysis revealed three types of food preferences: preference for fruits and vegetables, preference for a varied diet and low food preference. Gender, dietary knowledge, residence, education, and BMI were revealed as significant predictors of the class memberships. From baseline to follow-up, the most stable status in young adult group was preference for fruits and vegetables as it had a high transition probability 0.764 of remaining in the same class. While low food preference and preference for a varied diet statuses were unstable, they had transition probabilities of 0.590 and 0.554, respectively, moving to the preference for fruits and vegetables in year 2015. For middle-aged and elderly groups, the most stable and unstable statuses were preference for fruits and vegetables and preference for a varied diet, respectively. The unstable classes all had over 50% probability of moving to the preference for fruits and vegetables after four years. Potential explanations and implications are also discussed.     

The role of political connection on overinvestment of Chinese energy firms

This paper explores the role of political connection on the overinvestment of Chinese energy firms. We argue that political connection can act as a “helping hand” that enables energy firms to obtain more government support to invest and a “grabbing hand” that forces politically connected energy firms to heavily overinvest for the promotion benefit of local politicians. Our analysis shows that overinvestment of an energy firm is positively affected by its political connection. In terms of the “helping hand” effects, we find that politically connected energy firms are more likely overinvest when they receive more government subsidy. In terms of the “grabbing hand” effects, we show that local politician is more likely to force energy firms to overinvest if the politician is approaching the 65-year-old promotion line. Although firms are less likely to overinvest during the economic downturn, local politician shapes the investment decision of energy firms through the “grabbing hand” effects.     

Characteristic of filamentous fungal diversity and dynamics associated with wheat Qu and the traditional fermentation of Chinese rice wine

Chinese rice wine, a traditional alcoholic beverage, is fermented from steamed rice with a starter culture of wheat Qu which is either inoculated spontaneously with various microorganisms in natural environment or inoculated artificially with certain microorganisms. The diversity of filamentous fungi in wheat Qu and the fermentation mashes of Chinese rice wine was studied by high‐throughput sequencing. The results indicated that filamentous fungi varied in wheat Qu and fermentation periods. Moreover, Aspergillus was the dominant filamentous fungi at genus levels. On that basis, quantitative real‐time PCR was used to quantify some dominant Aspergillus species. It was found that Aspergillus species in wheat Qu were more abundant than that in various fermentation mashes and continually decreased until the end of rice wine fermentation. Furthermore, it was found that different organic acids were produced by microorganisms at different fermentation stages might due to the changes of oxygen and nutrient content.