[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fbjqr4ND_RBdT3VbrkgDg1Lh0P_wAxKI-QeK4DqWu4LM":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":8,"question":15,"related":16,"source":27,"type":112},[],"2025-06-23 23:13:36",207345320,[],{"count":9,"courseId":10,"courseImg":11,"courseName":12,"workId":13,"workName":14},48,"8918ca170ab255299558f67cd4aa8a6a","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fda81200f3c2573fd99e5376ca1429a86.jpg","机器学习","work_43205571","第6章 支持向量机","任何一个核函数, 都隐式地定义了一个[] (九个字)",[17,29,38,48,57,66,75,84,94,103],{"answer":18,"createTime":19,"id":20,"options":21,"question":26,"source":27,"type":28},[],"2025-06-24 00:41:26",185927369,[22,23,24,25],"对于线性可分的训练样本,存在唯一的超平面将训练样本全部分类正确","间隔只与 w 有关, 与 b 无关","支持向量机训练完成后, 最后的解与所有训练样本都有关","对于线性可分的训练样本, 支持向量机算法学习得到的能哆将训练样本正确分类且具有"最大间隔"的超平面是存在并且唯一的","机基本型中间隔、支持向量和超平面 w x+b=0 的说法,下列说法正确的是","v1",0,{"answer":30,"createTime":19,"id":31,"options":32,"question":37,"source":27,"type":28},[],185927371,[33,34,35,36],"在推导对偶问题时, 引入的拉格朗日乘子没有约束条件","对偶问题需要满足KKT条件","对偶问题的最优值是原始问题最优值的下界","通过对偶问题推导出的模型表达式能够体现解的稀疏性","下面关于支持向量机对偶问题的说法错误的是",{"answer":39,"createTime":40,"id":41,"options":42,"question":47,"source":27,"type":28},[],"2025-06-24 00:41:27",185927373,[43,44,45,46],"高斯核函数","0\u002F1损失函数","交叉熵函数","指数损失函数","以下属于常用核函数的是( )",{"answer":49,"createTime":19,"id":50,"options":51,"question":56,"source":27,"type":28},[],185927375,[52,53,54,55],"软间隔是支持向量机缓解过拟合的一种手段","采用hinge损失函数后仍保持了SVM解的稀疏性","正则化参数C越小, 模型对分类错误的容忍度越高","软间隔的基本思路为在最大化间隔的同时, 让不满足约束的样本尽可能少","下面有关软间隔支持向量机说法错误的是 ()",{"answer":58,"createTime":40,"id":59,"options":60,"question":65,"source":27,"type":28},[],185927377,[61,62,63,64],"应用核函数","回代求解参数","令偏导数为零","引入拉格朗日乘子","SVM对偶问题求解中,拉格朗日乘子法第二步是( )",{"answer":67,"createTime":19,"id":68,"options":69,"question":74,"source":27,"type":28},[],185927379,[70,71,72,73],"仅适用于线性可分数据","严格等于0\u002F1损失","是0\u002F1损失的上界","非凸且非连续","替代损失函数需满足的条件是( )",{"answer":76,"createTime":19,"id":77,"options":78,"question":83,"source":27,"type":28},[],185927381,[79,80,81,82],"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","以上选项存在说法错误的","支持向量机基本型是一个凸二次规划问题","下面关于支持向量机的说法错误的是",{"answer":85,"createTime":86,"id":87,"options":88,"question":93,"source":27,"type":28},[],"2025-06-23 23:13:35",185927383,[89,90,91,92],"高维空间的离散点","线性模型的组合","随机森林的集成","核函数的线性组合","表示定理表明,SVM模型可以表示为( )",{"answer":95,"createTime":40,"id":96,"options":97,"question":102,"source":27,"type":28},[],185927385,[98,99,100,101],"无约束优化问题","随机优化问题","有约束优化问题","非凸优化问题","拉格朗日乘子法用于解决( )",{"answer":104,"createTime":19,"id":105,"options":106,"question":111,"source":27,"type":28},[],185927387,[107,108,109,110],"支持向量回归模型的解仍然具有稀疏性","间隔带两侧的松弛程度可有所不同","支持向量回归也存在对偶问题","支持向量回归一般要求损失为 0 当且仅当模型的输出和实际值一样","下面关于支持向量回归, 说法错误的是",2]