A new reverse reduce-error ensemble pruning algorithm

Although greedy algorithms possess high efficiency, they often receive suboptimal solutions of the ensemble pruning problem, since their exploration areas are limited in large extent. And another marked defect of almost all the currently existing ensemble pruning algorithms, including greedy ones, consists in: they simply abandon all of the classifiers which fail in the competition of ensemble selection, causing a considerable waste of useful resources and information. Inspired by these observations, an interesting greedy Reverse Reduce-Error (RRE) pruning algorithm incorporated with the operation of subtraction is proposed in this work. The RRE algorithm makes the best of the defeated candidate networks in a way that, the Worst Single Model (WSM) is chosen, and then, its votes are subtracted from the votes made by those selected components within the pruned ensemble. The reason is because, for most cases, the WSM might make mistakes in its estimation for the test samples. And, different from the classical RE, the near-optimal solution is produced based on the pruned error of all the available sequential subensembles. Besides, the backfitting step of RE algorithm is replaced with the selection step of a WSM in RRE. Moreover, the problem of ties might be solved more naturally with RRE. Finally, soft voting approach is employed in the testing to RRE algorithm. The performances of RE and RRE algorithms, and two baseline methods, i.e., the method which selects the Best Single Model (BSM) in the initial ensemble, and the method which retains all member networks of the initial ensemble (ALL), are evaluated on seven benchmark classification tasks under different initial ensemble setups. The results of the empirical investigation show the superiority of RRE over the other three ensemble pruning algorithms.     

指纹细节特征提取与剪枝

指纹图像经预处理后，由于各种原因，会产生数以百计的特征。为了图像匹配的精确和快速，提出一套在细化后的指纹点线图上提取细节特征并对其剪枝的方法，并根据细节特征的拓扑特性，设计了验证其可靠性的算法。通过实验，证明了该后处理过程在去除指纹图像伪特征方法面具有良好的效果。     

Scheduling in or-parallel Prolog systems: Survey and open problems

Implementation of or-parallel Prolog systems offers a number of interesting scheduling problems. The main issues are the interaction between memory models and scheduling, ordering of multiple solutions, and scheduling of speculative work. The problems occur partly because of the design choices (e.g. the choice of a memory model), and partly because of the desire to maintain observational equivalence between parallel and sequential implementations of Prolog, while achieving high efficiency. In the first part of this paper a common framework for discussing scheduling in or-parallel systems is introduced, and also a collection of issues that must be addressed in such systems is presented. In the second part of the paper we survey a number of solutions to these problems comparing their efficiency whenever possible. We close the survey with a short discussion of open problems.Current association: Carlstedt Elektronik AB     

Aeromagnetic compensation using neural networks

Airborne magnetic surveys in geophysical exploration can be subject to interference effects from the aircraft. Principal sources are the permanent magnetism of various parts of the aircraft, induction effects created by the earth's magnetic field and eddy-current fields produced by the aircraft's manoeuvres. Neural networks can model these effects as functions of roll, pitch, heading and their time derivatives, together with vertical acceleration, charging currents to the generator, etc., without assuming an explicit physical model. Separation of interference effects from background regional and diurnal fields can also be achieved in a satisfactory way.     

基于灵敏度分析的FPGM剪枝算法研究

针对等比例剪枝导致的重要卷积层剪枝过度、残留大量冗余参数以及精度损失较大的问题,在FPGM剪枝策略基础上融入灵敏度分析进行网络剪枝。算法采用精度反馈来分析每一层卷积层的重要性,控制单层剪枝比例分析每一层不同剪枝比例对精度损失的影响,获取各个卷积层的灵敏度;结合FPGM策略分析卷积层内卷积核的重要程度按灵敏度的剪枝比例剪掉不重要的卷积核,达到对神经网络进行有效剪枝的目的。实验结果表明,所提方法在MobileNet-v1和ResNet50上剪枝率为50%的情况下,精确度仅下降1.56%和0.11%;所提方法在精度损失一致下,ResNet50上具有更高剪枝率和更低计算量。     

选择性集成算法分类与比较

选择性集成是当前机器学习领域的研究热点之一。由于选择性集成属于NP"难"问题,人们多利用启发式方法将选择性集成转化为其他问题来求得近似最优解,因为各种算法的出发点和描述角度各不相同,现有的大量选择性集成算法显得繁杂而没有规律。为便于研究人员迅速了解和应用本领域的最新进展,本文根据选择过程中核心策略的特征将选择性集成算法分为四类,即迭代优化法、排名法、分簇法、模式挖掘法;然后利用UCI数据库的20个常用数据集,从预测性能、选择时间、结果集成分类器大小三个方面对这些典型算法进行了实验比较;最后总结了各类方法的优缺点,并展望了选择性集成的未来研究重点。     

基于协方差特征的裁剪AdaBoost算法

针对Haar-like特征的缺陷以及AdaBoost算法存在训练耗时的问题,提出一种基于协方差特征的改进AdaBoost人脸检测算法。该方法用协方差特征代替Haar-like特征进行特征提取;然后使用特征裁剪和动态权重裁剪相结合的裁剪AdaBoost算法,训练得到基于协方差特征的强分类器。实验结果表明,相对于基于Haar-like特征的AdaBoost算法,该算法性能没有明显退化且很大程度上提高了训练速度。     

基于逐层剪枝的中文高频重复模式快速提取算法

为了从大规模语料中快速提取高频重复模式,以递增n-gram模型为基础,使用散列数据结构提取重复串,并提出了一种基于低频字符和层次剪枝的逐层剪枝算法,用于过滤低频垃圾字串,减少I/O读写次数。在此基础上,应用改进的字串排序算法,使字符串排序可在O(n)时间内完成,从而有效提高重复模式的提取效率。实验表明,该算法是一种有效的重复模式提取算法,其I/O读写次数同语料规模呈线性关系,远小于使用首字符进行语料划分的方法,能快速有效地从规模远大于内存容量的文本语料中提取重复模式,特别适合于大规模语料的高频重复模式提取,对以重复模式为基础的新词识别、术语抽取等具有重要的支撑作用。     

基于SuperPoint的轻量级特征点及描述子提取网络

图像特征点及描述子提取是SLAM、SFM和3D重建等任务的基础,较好的图像特征点及描述子提取算法会对这些任务的进步产生十分重要的作用.本文聚焦于提取特征点和描述子算法中鲁棒性较高、性能较好的SuperPoint网络,对该网络进行了一定程度的改进.针对其计算量和参数较大的问题,首先将普通卷积改成深度可分离卷积,改变卷积层数和下采样方式,之后改进通道剪枝算法,使其可以应用于深度可分离卷积,对网络进行剪枝.实验结果显示,在轻微损失特征点检测和匹配效果的情况下,将网络参数量压缩为原来网络的15％,运算量压缩为原来网络的5％,FPS提升6.64倍,取得了较好的实验效果.     

基于卷积神经网络的实时车辆检测

为了解决对于尺度变换较大车辆及遮挡车辆检测性能不足的问题,提出了一种实时车辆检测模型。针对车辆检测算法对于尺度敏感的问题,通过使用深度残差网络作为特征提取层,构建特征金字塔网络用于多尺度检测;利用软化非极大抑制线性衰减置信得分解决车辆遮挡问题,从而降低车辆的漏检率;同时对模型进行通道级裁剪缩减模型参数规模,节省计算资源,提高模型检测速度。在VOC数据集上进行实验,结果表明,提出的方法在检测精度和检测速度上均获得较高的性能。在检测精度上,达到87.6%的准确率,相较于YOLOv3提升了3.7个百分点,相较于SSD提升了9.8个百分点;在检测速度上,每秒检测帧数达到42 f/s,实现了车辆的实时检测。特别地,将模型应用于环境复杂的Apollo数据集,相较于YOLOv3具有更好的鲁棒性。