Probabilistic learning on manifolds (PLoM) with partition

The probabilistic learning on manifolds (PLoM) introduced in 2016 has solved difficult supervised problems for the “small data” limit where the number N of points in the training set is small. Many extensions have since been proposed, making it possible to deal with increasingly complex cases. However, the performance limit has been observed and explained for applications for which N is very small and for which the dimension of the diffusion-map basis is close to N. For these cases, we propose a novel extension based on the introduction of a partition in independent random vectors. We take advantage of this development to present improvements of the PLoM such as a simplified algorithm for constructing the diffusion-map basis and a new mathematical result for quantifying the concentration of the probability measure in terms of a probability upper bound. The analysis of the efficiency of this extension is presented through two applications.     

机器翻译译文质量估计综述

机器翻译译文质量估计(Quality Estimation,QE)是指在不需要人工参考译文的条件下,估计机器翻译系统产生的译文的质量,对机器翻译研究和应用具有很重要的价值。机器翻译译文质量估计经过最近几年的发展,取得了丰富的研究成果。该文首先介绍了机器翻译译文质量估计的背景与意义;然后详细介绍了句子级QE、单词级QE、文档级QE的具体任务目标、评价指标等内容,进一步概括了QE方法发展的三个阶段: 基于特征工程和机器学习的QE方法阶段,基于深度学习的QE方法阶段,融入预训练模型的QE方法阶段,并介绍了每一阶段中的代表性研究工作;最后分析了目前的研究现状及不足,并对未来QE方法的研究及发展方向进行了展望。     

快速近似计算Shapley值的归因解释方法

Shapley值归因解释方法虽然能更准确量化解释结果, 但过高的计算复杂度严重影响了该方法的实用性. 本文引入KD树重新整理待解释模型的预测数据, 通过在KD树上插入虚节点, 使之满足TreeSHAP算法的使用条件, 在此基础上提出了KDSHAP方法. 该方法解除了TreeSHAP算法仅能解释树结构模型的限制, 将该算法计算Shapley值的高效性放宽到对所有的黑盒模型的解释中, 同时保证了计算准确度. 通过实验对比分析, KDSHAP方法的可靠性, 以及在解释高维输入模型时的适用性.     

Automated prediction system of household energy consumption in cities using web crawler and optimized artificial intelligence

The supply of electrical energy is critical to convenient and comfortable living. However, people consume a large amount of energy, contributing to an energy crisis and global warming, and damaging some ecological cycles. Residential electricity consumption has greater elasticity than industrial and business consumption; it therefore has high energy-saving potential. This work establishes an automated platform, which provides information about residential electricity consumption in each city in Taiwan. Machine learning was used to forecast future residential electricity demand. A nature-inspired optimization method was applied to enhance the accuracy of the best machine learner, yielding an even better hybrid ensemble model. Performance measures indicate that the resulting model is accurate and provides effective information for reference. An automatic web-based system based on the model was combined with a web crawler and scheduled to run automatically to provide information on monthly residential electricity consumption in each county and city. By providing energy consumption information across the country, power providers and government can discuss policy and set different goals for energy use. The results of this study can facilitate the early implementation of energy-saving and carbon emission-reducing in cities and aid utility companies in establishing energy conservation guidelines.     

Prediction of fish quality level with machine learning

In this study, sea bream, sea bass, anchovy and trout were captured and recorded using a digital camera during refrigerated storage for 7 days. In addition, their total viable counts (TVC) were determined on a daily basis. Based on the TVC, each fish was classified as ‘fresh’ when it was <5 log cfu per g, and as ‘not fresh’ when it was >7 log cfu per g. They were uploaded on a web-based machine learning software called Teachable Machine (TM), which was trained about the pupils and heads of the fish. In addition, images of each species from different angles were uploaded to the software in order to ensure the recognition of fish species by TM. The data of the study indicated that the TM was able to distinguish fish species with high accuracy rates and achieved over 86% success in estimating the freshness of the fish species tested.     

草原石林：东北亚早白垩世花岗岩伸展穹隆和岩浆成矿爆发的地质遗迹及其形成过程

东北亚地区至少发育9个大型草原石林,属花岗岩石林的一种,是东北亚早白垩世巨量伸展和第四系冰川、风化作用的共同产物,主要特点是发育密集的近水平节理。本文通过整理岩浆岩专题填图和构造研究发现,这些石林的主体是早白垩世东北亚地区巨量伸展过程中形成众多的不对称花岗岩穹隆或同减薄岩体,岩体边部发育韧性剪切带,核部发布近水平岩浆流动面理,伸展和岩浆多期侵位导致与成矿密切相关的岩浆热液大量上涌,岩体抬升至地表,沿岩浆面理发育密集的近水平节理和部分垂直节理,后期沿这些构造面理发生的物理风化和化学作用(包括风蚀、冰川以及盐风化等)造就了千奇百怪的形态。草原石林是花岗岩地貌中的“另类”,不仅具有指示东北亚地区花岗岩形成时代(早白垩世)的特点,也是寻找晚中生代中国北方大规模成矿的良好标志体,而且风化形成的千姿百态的景色极具观赏价值,可以作为整个蒙古草原地区和大兴安岭南段地区特色的地学科普和生态旅游资源。     

智能数据分区与布局研究

大数据时代,数据规模庞大,由数据进行驱动的应用分析场景日益增多.如何快速、高效地从这些海量数据中提取出用以分析决策的信息,给数据库系统带来重大挑战.同时,现代商业分析决策对分析数据的实时性要求数据库系统能够同时快速处理ACID事务和复杂的分析查询.然而,传统的数据分区粒度太粗,且不能适应动态变化的复杂分析负载;传统的数据布局单一,不能应对现代大量增加的混合事务分析应用场景.为了解决以上问题,“智能数据分区与布局”成为当前的研究热点之一,它通过数据挖掘、机器学习等技术抽取工作负载的有效特征,设计最佳的分区策略来避免扫描大量不相关的数据,指导布局结构设计以适应不同类型的工作负载.首先介绍了智能数据分区与布局的相关背景知识,然后对智能数据分区与布局技术的研究动机、发展趋势、关键技术进行详细的阐述.最后,对智能数据分区与布局技术的研究前景做出总结与展望.     

Defect control and ionic conductivity of oxynitride perovskite Sr0.83Li0.17Ta0.83O1.88N0.74

Perovskite lattice was tailored by introducing site vacancies and mixed anion composition, to produce Sr_0.83Li_0.17Ta_0.83O_1.88N_0.74 (Li02N). Further, Li02N was converted to a defect oxide Sr_0.83Li_0.17Ta_0.83O₃ (Li02O) by applying an optimized treatment: heating in air at 1173 K for 2 h. According to the neutron Rietveld refinement, Li02N and Li02O are tetragonal and orthorhombic, respectively, where the lattice volume of Li02O is significantly smaller than that of Li02N. The ionic conductivity (σ_ion) of Li02N and Li02O was evaluated by the ac impedance spectroscopy and the equivalent circuit analysis. Both Li02N (σ_ion = 10^?5.5 S/cm at 671 K) and Li02O (σ_ion = 10^?6.2 S/cm at 667 K) exhibited an Arrhenius behavior of ionic conductivity with activation energies of 0.87 eV and 0.75 eV, respectively. It is interpreted that the nitride component enhances the ionic conduction of Li02N, while the vacancy of the anion lattice makes an opposite effect.     

In Vivo Albumin-Binding of a C-Functionalized Cyclam Platform for 64Cu-PET/CT Imaging in Breast Cancer Model

An improved glucose-chelator-albumin bioconjugate (GluCAB) derivative, GluCAB-2^Mal, has been synthesized and studied for in vivo ⁶⁴Cu-PET/CT imaging in breast cancer mice models together with its first-generation analogue GluCAB-1^Mal. The radioligand works on the principle of tumor targeting through the enhanced permeability and retention (EPR) effect with a supportive role played by glucose metabolism. [⁶⁴Cu]Cu-GluCAB-2^Mal (99 % RCP) exhibited high serum stability with immediate binding to serum proteins. In vivo experiments for comparison between tumor targeting of [⁶⁴Cu]Cu-GluCAB-2^Mal and previous-generation [⁶⁴Cu]Cu-GluCAB-1^Mal encompassed microPET/CT imaging and biodistribution analysis in an allograft E0771 breast cancer mouse model. Tumor uptake of [⁶⁴Cu]Cu-GluCAB-2^Mal was clearly evident with twice as much accumulation as compared to its predecessor and a tumor/muscle ratio of up to 5 after 24 h. Further comparison indicated a decrease in liver accumulation for [⁶⁴Cu]Cu-Glu-CAB-2^Mal.