一种混叠语音清浊音判断方法

提出了一种混叠语音清浊音判断方法。该方法在分频带自相关函数的混叠语音基频分离提取算法中,引入了双周期度及频带组的分布度两个参数用于对混叠语音清浊判断,从而使基于分频带自相关函数的混叠语音基频分离提取方法不再局限于浊音,可推广到清音信号。实验结果证明算法有一定的清浊音判断能力。     

Learning Strategies for Mechanised Building of Decision Procedures

In this paper we present an investigation into whether and how decision procedures can be learnt and built automatically. Our approach consists of two stages. First, a refined brute-force search procedure applies exhaustively a set of given elementary methods to try to solve a corpus of conjectures generated by a stochastic context-free grammar. The successful proof traces are saved. In the second stage, a learning algorithm (by Jamnik et al.) tries to extract a required supermethod (i.e., decision procedure) from the given traces. In the paper, this technique is applied to elementary methods that encode the operations of the Fourier-Motzkin's decision procedure for Presburger arithmetic on rational numbers. The results of our experiment are encouraging.     

自动售货机集群管理专家决策系统

主要讨论了专家决策系统在自动售货机集群管理中的应用,给出了一个通过无线网络联结的自动售货机集群的专家决策系统的系统结构,以及构造该系统的步骤过程,建立了自动售货机集群管理专家决策系统的模型库,利用决策模型对所采集的自动售货机中的数据进行分析,并根据专家系统的推理得出专家结论,给操作人员一个决策的参考,实验证明自动售货机专家决策系统的应用降低了售货机的运营管理成本且提高了运营效率。     

Knowledge-Based Adaptive Decision Tree State Tying for Conversational Speech Recognition

This paper presents a new method of constructing phonetic decision trees (PDTs) for acoustic model state tying based on implicitly induced prior knowledge. Our hypothesis is that knowledge on pronunciation variation in spontaneous, conversational speech contained in a relatively large corpus can be used for building domain-specific or speaker-dependent PDTs. In view of tree-structure adaptation, this method leads to transformation of tree topology in contrast to keeping fixed tree structure as in traditional methods of speaker adaptation. A Bayesian learning framework is proposed to incorporate prior knowledge on decision rules in a greedy search of new decision trees, where the prior is generated by a decision tree growing process on a large data set. Experimental results on the telemedicine automatic captioning task demonstrate that the proposed approach results in consistent improvement in model quality and recognition accuracy.     

无逆矩阵在线序列极限学习机*

无逆矩阵极限学习机只能以批量学习方式进行训练,将其拓展为无逆矩阵在线学习版本,提出了无逆矩阵在线序列极限学习机算法(IOS-ELM).所提算法增加训练样本时,利用Sherman Morrison Woodbury公式对新增样本数据后的模型进行更新,直接计算出新增隐含层输出权重,避免对已经分析过的训练样本的输出权重进行重...     

Weighted Online Sequential Extreme Learning Machine for Class Imbalance Learning

Most of the existing sequential learning methods for class imbalance learn data in chunks. In this paper, we propose a weighted online sequential extreme learning machine (WOS-ELM) algorithm for class imbalance learning (CIL). WOS-ELM is a general online learning method that alleviates the class imbalance problem in both chunk-by-chunk and one-by-one learning. One of the new features of WOS-ELM is that an appropriate weight setting for CIL is selected in a computationally efficient manner. In one-by-one learning of WOS-ELM, a new sample can update the classification model without waiting for a chunk to be completed. Extensive empirical evaluations on 15 imbalanced datasets show that WOS-ELM obtains comparable or better classification performance than competing methods. The computational time of WOS-ELM is also found to be lower than that of the competing CIL methods.     

智能机器伦理决策设计研究综述*

为了解决机器伦理决策困境,分析智能机器伦理决策设计的研究现状,对机器伦理决策设计方法进行分类总结。对基于规则推理和基于神经科学不同的伦理决策设计实现方式及优缺点进行了对比分析,并对现阶段无人驾驶伦理决策研究现状进行了归纳分析,最后总结了现阶段机器伦理决策的研究水平。基于上述分析研究,总结得出了解决机器伦理决策困境的技术难点有:机器伦理道德地位的确立、普适机器伦理决策、机器伦理决策评估。提出了未来研究方向:将人类道德规范应用到机器伦理决策设计中,运用机器学习实现伦理决策设计,利用法律及行业标准规范来实现机器伦理决策设计。为机器伦理决策及无人驾驶伦理决策的设计提供参考。     

Augmented Online Sequential Quaternion Extreme Learning Machine

Neural Processing Letters - Online sequential extreme learning machine (OS-ELM) is one of the most popular real-time learning strategy for feedforward neural networks with single hidden layer due...     

Virtual Machine Communication for the Implementation of Decision Support Systems

This paper presents the use of the virtual machine concept as a software engineering tool. The focus is on techniques that allow the rapid integration and assimilation of existing data, models, analytical facilities, report generation facilities, and data base management systems. Since many of these models, programs, and facilities run under different operating systems, they often are incompatible with each other. By combining multiple virtual machines in a particular configuration, and allowing for communication among them, it has been possible to overcome these difficulties. The result is a set of software engineering tools that seem particularly useful in decision support systems. We present the application of these tools along with the software techniques used to implement them, and quantify some of their costs and benefits.     

机器学习的可解释性

近年来,机器学习发展迅速,尤其是深度学习在图像、声音、自然语言处理等领域取得卓越成效.机器学习算法的表示能力大幅度提高,但是伴随着模型复杂度的增加,机器学习算法的可解释性越差,至今,机器学习的可解释性依旧是个难题.通过算法训练出的模型被看作成黑盒子,严重阻碍了机器学习在某些特定领域的使用,譬如医学、金融等领域.目前针对机器学习的可解释性综述性的工作极少,因此,将现有的可解释方法进行归类描述和分析比较,一方面对可解释性的定义、度量进行阐述,另一方面针对可解释对象的不同,从模型的解释、预测结果的解释和模仿者模型的解释3个方面,总结和分析各种机器学习可解释技术,并讨论了机器学习可解释方法面临的挑战和机遇以及未来的可能发展方向.     

板坯连铸机顺序控制分析

介绍了顺序控制在板坯连铸机生产中的应用,分析了铸机的运行状态转换及启动过程.该系统采用N-90分布式控制系统,将每个控制对象的整个复杂动作过程分解为符合工艺要求的具体步骤,软件设计由十分丰富的标准化模块功能码和图形库组态而成,编程灵活、实时性强、可扩展性好、能实现严格的逻辑顺序控制功能.该系统投运以来,控制准确,运行稳定可靠,保障了设备安全使用,在整个连铸控制中起到了重要作用.     

UCAV对地多目标序次攻击决策研究

在增强无人机战斗力的研究中,针对无人作战飞机(UCAV)对地多目标序次攻击决策问题展开研究,建立多目标序次攻击的决策模型.为提高UCAV执行多目标攻击任务的成功率,决策模型综合考虑了UCAV在整个任务过程中所面临的战场风险,并将决策算法与航路规划问题分解,提出一种基于威胁分布密度的飞机生存概率评估方法,提高了决策效率.将序次决策问题转化为最优搜索问题,利用自适应蚁群算法求解决策问题并给出具体算法流程.仿真结果表明,多目标序次攻击决策方法能有效解决复杂战场环境分布所带来的目标排序优化问题,给出合理的多目标攻击时序,为方案设计提供了依据.     

Feature Adaptive Online Sequential Extreme Learning Machine for lifelong indoor localization

Wi-Fi-based indoor localization with high capability and feasibility needs to implement lifelong online learning mechanism. However, the characteristic of Wi-Fi is wide variability, which lies in not only the fluctuation of signal strength value, but also the increase or decrease in the number of access points (APs). The traditional algorithms are effective for signal fluctuation, but cannot handle the dimension-changing problem of features caused by increase and decrease in APs’ number. To solve this problem, we propose a Feature Adaptive Online Sequential Extreme Learning Machine (FA-OSELM) algorithm. It can transfer the original model to a new one with a small number of data with new features, so as to make the new model suitable for the new feature dimension. The experiments show that the FA-OSELM can get higher accuracy with a small amount of new data, and it is an effective method to make lifelong indoor localization practical.     

语音通信中信号的清浊音判别

文章采用了一种基于三参数组合的方法对语音信号进行清／浊音判决。在传统两参数（短时能量和过零率）算法的基础上增加了自相关函数作为判决参数,减少了清／浊音信号的误判率。经计算机仿真表明,算法切实可行,且算法误判概率小。     

一种基于决策树模型的音库构建和基元选取方法

针对嵌入式设备的存储容量小、计算能力有限的特点,设计了一种基于CART（Classification and Regression Trees）决策树模型的基元预选算法和基元选取算法,可以从原始语音语料库中挑选出最有代表性的基元样本,从而有效地降低音库规模和算法的复杂度,满足了嵌入式TFS（Text-to-Speech）系统的需要。基于以上算法,移动终端上实现了一个嵌入式中文TTS系统,实验结果表明该系统的合成语音具有较高的可懂度和自然度。     

Learning Sequential Decision Rules Using Simulation Models and Competition

The problem of learning decision rules for sequential tasks is addressed, focusing on the problem of learning tactical decision rules from a simple flight simulator. The learning method relies on the notion of competition and employs genetic algorithms to search the space of decision policies. Several experiments are presented that address issues arising from differences between the simulation model on which learning occurs and the target environment on which the decision rules are ultimately tested.     

医学图像决策支持系统中的SVM算法

支持向量机(SVM)方法是利用最优分类面(线)将两类样本在特征空间或输入空间中无错误地分开,而且要使两类的分类空隙最大。因此标准的SVM方法需要求解二次规划问题,计算量很大。该文以一个医学决策支持系统为应用背景,介绍一种解决该问题的新方法。在UCI数据集和所开发的决策支持系统上的应用表明,该算法简便可行,具有更高的精度和更快的速度。     

Random Forests of Phonetic Decision Trees for Acoustic Modeling in Conversational Speech Recognition

In this paper, we present a novel technique of constructing phonetic decision trees (PDTs) for acoustic modeling in conversational speech recognition. We use random forests (RFs) to train a set of PDTs for each phone state unit and obtain multiple acoustic models accordingly. We investigate several methods of combining acoustic scores from the multiple models, including maximum-likelihood estimation of the weights of different acoustic models from training data, as well as using confidence score of -value or relative entropy to obtain the weights dynamically from online data. Since computing acoustic scores from the multiple models slows down decoding search, we propose clustering methods to compact the RF-generated acoustic models. The conventional concept of PDT-based state tying is extended to RF-based state tying. On each RF tied state, we cluster the Gaussian density functions (GDFs) from multiple acoustic models into classes and compute a prototype for each class to represent the original GDFs. In this way, the number of GDFs in each RF tied state is decreased greatly, which significantly reduces the time for computing acoustic scores. Experimental results on a telemedicine automatic captioning task demonstrate that the proposed RF-PDT technique leads to significant improvements in word recognition accuracy.     

一种高噪声环境下语音检测算法的仿真与应用

该文提出了一种高噪声环境下的自适应语音检测新算法。该算法利用了语音短时能量及帧内短时自相关的特性,两个自适应判决门限根据期望的误判率调整,无需事先给出噪声统计信息,适用于平稳噪声、缓变的非平稳噪声及脉冲噪声的情况,且可以实时实现。     

多目标分散决策的目标规划模型和玻尔兹曼机解法

研究建立了分散组织结构下多目标分散决策的目标规划模型，根据随机神经网络──玻尔兹曼机的基本原理，提出了求解该问题的一种新方法。示例的仿真结果表明该方法是非常有效的。