[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fny2VpmSqgLE5NPuOjtmthK-hCvRgCLKclnVUaQcs0XQ":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":18,"related":19,"source":30,"type":113},[],"2025-05-19 14:35:32",189012169,[8,9,10,11],"所有样本的权重和保持不变","前一个基学习器分错的样本会获得更大的权重","只要权重调整的方向正确, Boosting算法的性能就可以获得理论保证","决策树可以直接处理带权重的样本",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":16},85,"fc0c1d00e84a5bdd920ca3eef531981e","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Ffb47e65ee6bbb8f896e6b4fca1783b07.png","机器学习初步","exam_156490776","下列关于Boosting算法中样本权重调整的说法中正确的是",[20,32,41,50,59,67,76,86,95,104],{"answer":21,"createTime":22,"id":23,"options":24,"question":29,"source":30,"type":31},[],"2025-05-19 14:35:30",189012061,[25,26,27,28],"训练样本个数大于属性可能的取值数","训练样本个数小于属性可能的取值数","训练样本不充分、属性可能的取值数有限","训练样本不充分、属性有无限种可能的取值","拉普拉斯修正不适用于以下哪种情况","v1",0,{"answer":33,"createTime":22,"id":34,"options":35,"question":40,"source":30,"type":31},[],189012063,[36,37,38,39],"自顶向下","自底向上","随机顺序","以上答案都不对","后剪枝过程在生成完全决策树后,()对树中所有非叶结点进行考察",{"answer":42,"createTime":22,"id":43,"options":44,"question":49,"source":30,"type":31},[],189012065,[45,46,47,48],"特征的联合概率","每个类别的后验概率","各类别的先验概率","特征之间的条件独立性","在朴素贝叶斯分类器中,贝叶斯定理主要用来计算什么",{"answer":51,"createTime":22,"id":52,"options":53,"question":58,"source":30,"type":31},[],189012068,[54,55,56,57],"非负性","同一性","稳定性","直递性","距离度量需满足的基本性质不包括()",{"answer":60,"createTime":22,"id":61,"options":62,"question":66,"source":30,"type":31},[],189012070,[63,64,65],"在正负类样本"正中间"的","靠近正类样本的","靠近负类样本的","对于线性可分的二分类任务样本集, 将训练样本分开的超平面有很多, 支持向量机试图寻找满足什么条件的超平面",{"answer":68,"createTime":22,"id":69,"options":70,"question":75,"source":30,"type":31},[],189012072,[71,72,73,74],"留出法","交叉验证法","自助法","以上三个选项都可以","下列什么方法可以用来获得从原始数据集中划分出 "测试集"",{"answer":77,"createTime":78,"id":79,"options":80,"question":85,"source":30,"type":31},[],"2025-05-19 14:35:31",189012075,[81,82,83,84],"通过任意支持向量都能够求解出截距项","为了提高鲁棒性, 通常使用所有支持向量求解的平均值","通过任意样本都能够求解出截距项","截距项的求解能够体现支持向量机学习到的超平面仅与少量支持向量有关","在求解支持向量机截距项的时候错误的说法是()",{"answer":87,"createTime":78,"id":88,"options":89,"question":94,"source":30,"type":31},[],189012077,[90,91,92,93],"训练集;需要","训练集;不需要","验证集;需要","验证集;不需要","我们通常将数据集划分为训练集,验证集和测试集进行模型的训练,参数的验证需要在()上进行,参数确定后()重新训练模型",{"answer":96,"createTime":78,"id":97,"options":98,"question":103,"source":30,"type":31},[],189012079,[99,100,101,102],"经验","专家","规则","实践","机器学习的经典定义是:利用 () 改善系统自身的性能",{"answer":105,"createTime":78,"id":106,"options":107,"question":112,"source":30,"type":31},[],189012082,[108,109,110,111],"泛化误差","经验误差","学习误差","训练误差","模型在"未来"样本上的误差被称作()",1]