[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f3F-FatVEYQjDDEqwEOUXNCWAufgNcQUspvcpBRb7TqE":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":18,"related":19,"source":23,"type":24},[],"2023-05-08 00:28:38",33262301,[8,9,10,11],"机器学习的研究目标是研究通用的学习算法、认知模型以及解决实际的工程问题","机械学习是指机器学习过程无需任何推理或其它的知识转换,直接吸取环境所提供的信息,是最简单、最基础的学习过程","演绎学习主要采用演绎推理的推理形式进行机器学习","示例学习是通过目标对象与源对象的相似性,从而运用源对象的求解方法来解决目标对象的问题",{"courseId":13,"courseImg":14,"courseName":15,"workId":16,"workName":17},"1000008418","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F8e391dadee7a948b8ae0df014f53e38f.jpg","人工智能原理与方法","22068511","人工智能原理与方法教程考试","机器学习是一门研究使用计算机获取新的知识和技能,提高现有计算机求解问题能力从而具有一定智能的科学.下列关于机器学习的说法中,错误的是( )",[20,25,35,44,52,57,62,67,76,85],{"answer":21,"createTime":5,"id":6,"options":22,"question":18,"source":23,"type":24},[],[8,9,10,11],"v2",0,{"answer":26,"createTime":5,"id":27,"options":28,"question":33,"source":23,"type":34},[],33262310,[29,30,31,32],"全连接层的作用就是分类,全连接层的每一个节点都与上一层的所有节点相连接,将前一层的特征进行进一步组合","卷积神经网络前面的各层是将图像的低维特征映射到抽象特征空间,全连接层则将抽象特征映射到样本标记空间中","损失函数的作用是用来估算模型的预测值 与实际值 的差距的函数","卷积神经网络中的池化层又称为下采样层,该层的目的是降低特征图的维度,进而降低卷积神经网络的参数量,减少过拟合","对于卷积网络来说,以下说法正确的是:( )",1,{"answer":36,"createTime":5,"id":37,"options":38,"question":43,"source":23,"type":34},[],33262315,[39,40,41,42],"宏观","微观","介观","三观","理解大脑包括哪几个层面:( )",{"answer":45,"createTime":5,"id":46,"options":47,"question":50,"source":23,"type":51},[],33262321,[48,49],"对","错","朴素贝叶斯可适用于非常小的数据集",3,{"answer":53,"createTime":5,"id":54,"options":55,"question":56,"source":23,"type":51},[],33262338,[48,49],"Q-Learning是强化学习算法中基于价值函数(value-based)的一种算法,主要思想就是将State与Action构建成一张Q值表(Q-table)来存储Q值,然后根据Q值来选取能够获得最大收益的动作",{"answer":58,"createTime":5,"id":59,"options":60,"question":61,"source":23,"type":51},[],33262339,[48,49],"粒子群算法即(POS算法)进行路径规划时,其流程是:初始化—>求个体机制与全局最优方向—>更新速度和位置—>终止条件判断",{"answer":63,"createTime":5,"id":64,"options":65,"question":66,"source":23,"type":51},[],33262353,[48,49],"问题求解技术主要包括两个方面,即问题的表示和求解的方法.问题的状态空间是一个表示该问题全部可能状态及其关系的图,它包含三种说明的集合",{"answer":68,"createTime":5,"id":69,"options":70,"question":75,"source":23,"type":34},[],33262356,[71,72,73,74],"神经元之间的联接强度决定信号传递的强弱","神经元之间的联接强度是可以随训练改变的","一个神经元接受的信号的累积效果决定该神经元的状态","信号可以是起刺激作用的,也可以是起抑制作用的","下列关于人工神经网络的基本特征说法正确的有( )",{"answer":77,"createTime":5,"id":78,"options":79,"question":84,"source":23,"type":34},[],33262370,[80,81,82,83],"按照不同的学习理论划分,机器学习模型可以分为有监督学习、半监督学习、无监督学习以及强化学习等不同类型","在实际应用中,机器学习主要以无监督学习或半监督学习方式为主","监督学习的监督体现在所有机器要处理的数据实现都要由人为定义好相应的类别,再对分类算法进行训练,最后得到可以使用的分类器","有监督学习的数据集有标签,无监督相比于有监督,没有训练过程,而是直接拿数据进行建模分析","下面关于机器学习的说法中,正确的有( )",{"answer":86,"createTime":5,"id":87,"options":88,"question":93,"source":23,"type":24},[],33262390,[89,90,91,92],"实现对于三维景物世界的理解,即实现人的视觉系统的某些功能","根据一幅或多幅二维图像计算出观测点到目标物体的距离","根据一幅或多幅二维图像计算出目标物体的运动参数","根据一幅或多幅二维图像计算出目标物体的表面物理特性","下列选项中,不属于计算机视觉需要达到的基本目的的是( )"]