[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fLffGZ06vaZUX0USGiMdftRxysvVNxanwX8--asu58rs":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":11,"question":17,"related":18,"source":29,"type":30},[],"2025-05-06 08:14:13",185805539,[8,9,10],"在正负类样本"正中间"的","靠近正类样本的","靠近负类样本的",{"count":12,"courseId":13,"courseImg":14,"courseName":15,"workId":16,"workName":15},85,"fc0c1d00e84a5bdd920ca3eef531981e","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Ffb47e65ee6bbb8f896e6b4fca1783b07.png","机器学习初步","exam_155400821","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",[19,31,40,49,58,67,76,85,94,101],{"answer":20,"createTime":21,"id":22,"options":23,"question":28,"source":29,"type":30},[],"2025-05-06 08:14:12",185805448,[24,25,26,27],"以很低概率得到不好的模型","以很低概率得到很好的模型","以很高概率得到很好的模型","以很高概率得到不好的模型","以下哪个是对概率近似正确(PAC)的正确解释","v1",0,{"answer":32,"createTime":21,"id":33,"options":34,"question":39,"source":29,"type":30},[],185805450,[35,36,37,38],"均方误差为0","均方误差最大","均方误差最小","均方误差为1","最小二乘参数估计得到的线性回归模型满足什么性质",{"answer":41,"createTime":5,"id":42,"options":43,"question":48,"source":29,"type":30},[],185805452,[44,45,46,47],"属性","标记","数据","学到的模型","以下哪个选项指的是机器学习中的假设(hypothesis)",{"answer":50,"createTime":5,"id":51,"options":52,"question":57,"source":29,"type":30},[],185805454,[53,54,55,56],"具有中心对称性","严格大于0","单调且任意阶可导","不需要写成分段形式","对数几率函数作为单位阶跃函数的替代函数的优点是什么",{"answer":59,"createTime":5,"id":60,"options":61,"question":66,"source":29,"type":30},[],185805456,[62,63,64,65],"支持向量机基本型是一个凸二次规划问题","将训练样本分开的超平面仅由支持向量决定","支持向量机的核心思想是最大化间隔","以上选项都存在说法错误","下面关于支持向量机的说法错误的是",{"answer":68,"createTime":5,"id":69,"options":70,"question":75,"source":29,"type":30},[],185805458,[71,72,73,74],"异质集成学习能获得更多样的个体","同质集成学习在实现上更加简单","同质集成学习能更方便的对比个体的结果","同质集成学习能获得性能更好的个体","下列关于同质集成学习与异质集成学习对比的说法中错误的是哪个",{"answer":77,"createTime":5,"id":78,"options":79,"question":84,"source":29,"type":30},[],185805460,[80,81,82,83],"纯度","对称差","大小","重要性","信息熵是度量样本集合()最常用的一种指标",{"answer":86,"createTime":5,"id":87,"options":88,"question":93,"source":29,"type":30},[],185805463,[89,90,91,92],"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":95,"createTime":5,"id":96,"options":97,"question":100,"source":29,"type":30},[],185805465,[46,98,45,99],"某种类型假设","运行速度","归纳偏好指机器学习算法在学习过程中,对以下哪个选项的偏好",{"answer":102,"createTime":5,"id":103,"options":104,"question":109,"source":29,"type":30},[],185805467,[105,106,107,108],"对偶问题需要满足KKT条件","通过对偶问题推导出的模型表达式能够体现解的稀疏性","在推导对偶问题时, 引入的拉格朗日乘子没有约束条件","对偶问题的最优值是原始问题最优值的下界","下面关于支持向量机对偶问题的说法错误的是()"]