Object detection based on visual memory: a feature learning and feature imagination process

ABSTRACT

Visual memory plays an important role for the human’s visual system to detect objects. The features of an object stored in the visual memory have much lower dimensions than the features contained within an image. We simulate the visual memory as a feature learning and feature imagination (FLFI) process to build an object detection algorithm. The method is constructed by a bottom-up feature learning and a top-down feature imagination. The proposed object detection method is tested using publicly available benchmark data sets, and the result indicates that it is fast and more robust.     

Is external research assessment associated with convergence or divergence of research quality across universities and disciplines? Evidence from the PBRF process in New Zealand

ABSTRACT

Performance-based research evaluations have been adopted in several countries both to measure research quality in higher education institutions and as a basis for the allocation of funding across institutions. Much attention has been given to evaluating whether such schemes have increased the quality and quantity of research. This paper examines whether the introduction of the New Zealand Performance-Based Research Fund process produced convergence or divergence in measured research quality across universities and disciplines between the 2003 and 2012 assessments. Two convergence measures are obtained. One, referred to as β-convergence, relates to the relationship between changes in average quality and the initial quality level. The second concept, referred to as σ-convergence, relates to the changes in the dispersion in average research quality over time. Average quality scores by discipline and university were obtained from individual researcher data, revealing substantial β- and σ-convergence in research quality over the period. The hypothesis of uniform rates of convergence across almost all universities and disciplines is supported. The results provide insights into the incentives created by performance-based funding schemes.     

Depreciation of Business R&D Capital

We develop a forward-looking profit model to estimate the depreciation rates of business R&D capital. By using U.S. government data between 1987 and 2007, and the newly developed model, we estimate both constant and time-varying industry-specific R&D depreciation rates. The results comprise a set of R&D depreciation rates for major U.S. high-tech industries. They align with the main conclusions from recent studies that the rates are in general higher than the traditionally assumed 15 percent and vary across industries. The relative ranking of the constant R&D depreciation rates among industries is consistent with industry observations and the industry-specific time-varying rates are informative about the dynamics of technological change and the levels of competition across industries. Lastly, we also present a cross-country comparison of the R&D depreciation rates between the U.S. and Japan, and find that the results reflect the relative technological competitiveness in key industries.     

基于灰色关联分析的公路工程造价估算研究

目前分析公路工程造价相似性问题过多依赖于主观经验判断法,缺乏有效合理的定量方法,本文利用灰色关联度分析原理,提出一种分析公路工程造价相似性的定量方法。首先选取相似案例库,采用熵值法求出该案例库的重要属性特征权重值,再计算出相似案例库与待估案例的灰色关联度,最后得出待估案例的公路工程造价估算值,并通过实例证明该估算模型研究的可行性。     

How to optimize the allocation of research resources? An empirical study based on output and substitution elasticities of universities in Chinese provincial level

Although significant progress has been made in China's basic research in recent years, there remains a wide gap between research in China and that from developed countries. How to optimize the allocative efficiency of research resources is of great importance for increasing research output. In this paper, using the fixed effect stochastic frontier model based on the translog production function, we estimate output and substitution elasticities of research and development (R&D) inputs at universities in China's provincial level during 2009–2016. We find that the R&D technical efficiency of China's universities, after a rapid growth, has tended to become relatively stable. Improvements of internationalization degree and exogenous R&D capabilities are conducive to promoting R&D technical efficiency, whereas expenditures from government grants inhibit the promotion of R&D technical efficiency; the effects of R&D capital deepening and internet penetration are not evident. The output elasticity of R&D capital is much higher than that of R&D personnel, suggesting that R&D capital is the main driving force of research output. The substitution elasticity between R&D capital and personnel has experienced a change from substitution to complementary since 2014. To realize sustained growth of research output, we should increase R&D input with positive output elasticity or reduce R&D input with negative output elasticity, making the necessary trade-offs according to the substitution relationship between the two R&D inputs.     

国外生态修复研究进展评述

研究目的:系统总结国外生态修复研究的重点进展和主流共识,为把握未来生态修复研究的前沿态势提供参考。研究方法:文献研究法。研究结果:按照生态修复的"目标导向—技术方法—成效评价"的框架进行梳理,现有研究在生态修复的对象及尺度、生态学及社会经济文化目标、各类型生态系统及要素的修复技术方法、成效评价维度及指标等方面取得了一定的研究进展,但总体而言,目前生态修复研究存在理论—实践—政策之间知识转移欠缺、生态学与社会经济学科交叉研究不足、全球气候变化背景下的研究缺乏等问题。研究结论:理论—实践—政策耦合研究、生态学—社会经济学科交叉研究、气候变化—生态修复联动研究等可能是未来生态修复领域的重点研究方向。     

深度学习技术在海关风险甄别中的应用研究

随着互联网的高速发展和人工智能时代的到来,越来越多从前必须由人脑完成的工作能够利用计算机技术来完成,而深度学习的出现更解决了传统机器学习算法在计算机视觉领域、自然语言处理领域表现不佳的问题,使机器也能够拥有准确感知图像和语音的能力。人脸识别是深度学习网络最常见的应用场景之一,具有自然、直接、方便的特点,且不需要检测对象配合,因此非常适合用于公共安全领域的风险检测。研究充分结合海关实际需求,搭建基于深度学习技术的人脸识别模型,提供对通关旅客进行实时风险甄别的解决方案,以及海关通关风险防控场景的理论参考,为后续深度学习技术在海关业务的研究提供支撑。     

海南自由贸易港海关通关监管模式研究

习近平总书记在庆祝海南建省办经济特区30周年大会上重要讲话和中央12号文件1精神,对海南建设中国特色自由贸易港作出了重要部署,也为海关支持海南建设中国特色自由贸易港提出了新的任务要求。本文针对这一背景,结合海关贯彻落实党的十九届四中全会精神部署要求,聚焦跨境贸易更加便利、风险防控更加精准、区域经济更加活跃的自贸港建设目标,研究提出"一线放开、二线高效管住"海关通关监管模式,并就如何强化保障、防控风险提出相应建议,积极助推海南建设中国特色自由贸易港。     

Structural breaks in the mean of dividend-price ratios: Implications of learning on stock return predictability

In their out-of-sample predictions of stock returns in the presence of structural breaks, Lettau and Van Nieuwerburgh (2008) implicitly assume that economic agents’ perception of the regime-specific mean for the dividend-price ratio is time-invariant within a regime. In this paper, we challenge this assumption and employ least squares learning with constant gain (or constant-gain learning) in estimating economic agents’ time-varying perception for the mean of dividend-price ratio. We obtain better out-of-sample predictions of stock returns than in Lettau and Van Nieuwerburgh (2008) for both the U.S. and Japanese stock markets. Our empirical results suggest that economic agents’ learning plays an important role in the dynamics of stock returns.     

Big data research in hospitality: From streetlight empiricism research to theory laden research

There is little doubt that Big Data is taking the center stage in academic research. For such reason this Editorial Viewpoint is intended to help authors drive a research agenda that is not only innovative but grounded in theory. An opportunity exists to demystify the view that with vast amount data the numbers will speak for themselves. Researchers can assist by creating synergies between Theory-driven and Data-driven research, especially since theory should play a critical role in driving critical dialogue about generating new knowledge. The goal of hospitality research with Big Data should be to create a virtuous research cycle between theory and data that generates knowledge and provide a better understanding that advances a more robust hospitality theory. This will help us rethink and inquire what constitutes knowledge in order to better understand the past, present, and future of hospitality research.