[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$ft3C6kKOhC_ZffjVeG-kEFv3K5VXpsgOa_Rk2e5hbajQ":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":15,"related":16,"source":24,"type":46},[],"2024-11-25 08:37:17",999757392,[8,9,10,11],"预测是计算信念状态,使用前向算法","平滑是给定已知证据序列,计算过去某一状态的后验概率分布过程,使用前向后向算法","滤波是给定已知证据序列,计算未来状态的后验概率分布,通过前向算法和转移模型计算","最可能解释是找到最可能产生观测结果的状态序列,可以使用维特比算法",{"courseImg":13,"courseName":14},"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fcf3bb414b5ea2367f316b2d3561124c7.jpg","[共享课]人工智能","下列说法正确的有( )",[17,26,36,47,52,61,70,75,78,87],{"answer":18,"createTime":5,"id":19,"options":20,"question":23,"source":24,"type":25},[],999757131,[21,22],"对","错","广度优先搜索可以找到步数最短的搜索路径,并且能保证路径的代价最小.( )","v2",3,{"answer":27,"createTime":5,"id":28,"options":29,"question":34,"source":24,"type":35},[],999757169,[30,31,32,33],"\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F337926b18a7ceaabdfad5b2639b7f157.jpg\">","\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F4380e14a56df3bb7de25cefb3358a2f9.jpg\">","\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fc960c31c4270d294fcb0f674bb6fc0af.jpg\">","\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F5f869f433b69cf3430fc9bb56d268ccd.jpg\">","在强化学习值函数近似中,蒙特卡罗方法对参数的更新公式是( )",0,{"answer":37,"createTime":5,"id":38,"options":39,"question":45,"source":24,"type":46},[],999757203,[40,41,42,43,44],"有限状态集合S","有限动作集合A","状态转移函数P","奖励函数R","衰减因子\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F942742265767fda821bd734bdc136d44.png\">","在无模型设定的强化学习中,马尔可夫决策过程的五元组已知部分有( )",1,{"answer":48,"createTime":5,"id":49,"options":50,"question":51,"source":24,"type":25},[],999757239,[21,22],"在alpha-beta剪枝算法中,对于MAX节点,当它的效用值比当前的alpha低时可以进行剪枝.( )",{"answer":53,"createTime":5,"id":54,"options":55,"question":60,"source":24,"type":46},[],999757271,[56,57,58,59],"如果启发式函数是可采纳的,则A*树搜索是最优的","UCS是A*算法的一个特殊情况","如果启发式函数是一致的,则A*图搜索是最优的","UCS图搜索和树搜索都是最优的","以下关于启发式函数和A*算法的描述正确的是( )",{"answer":62,"createTime":5,"id":63,"options":64,"question":69,"source":24,"type":35},[],999757328,[65,66,67,68],"更新=步长*预测误差","更新=步长*预测误差+特征值","更新=步长*预测误差*特征值","更新=步长*特征值+预测误差","在有模型的强化学习中,属于动态规划求解的是( )",{"answer":71,"createTime":5,"id":72,"options":73,"question":74,"source":24,"type":25},[],999757348,[21,22],"蒙特卡洛方法是基于转移\u003Cimg src=\"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fd14c6ffa1ca1cd5b96b8bc3ad047453c.png\">进行评估的.( )",{"answer":76,"createTime":5,"id":6,"options":77,"question":15,"source":24,"type":46},[],[8,9,10,11],{"answer":79,"createTime":5,"id":80,"options":81,"question":86,"source":24,"type":35},[],999757412,[82,83,84,85],"O(TX^2),O(TX)","O(TM^2),O(TM)","O(TMX),O(TX)","O(TMX),O(TM)","若转移矩阵是一个稀疏矩阵,且任何一个隐藏状态只能转移到M个可能的状态,使用维特比算法求最可能状态序列时可以忽略那些转移概率为0的路径,这时时间复杂度和空间复杂度为( )",{"answer":88,"createTime":5,"id":89,"options":90,"question":95,"source":24,"type":35},[],999757422,[91,92,93,94],"获胜状态的效用应该高于平局","可以通过特征线性组合的方式来计算","计算时间消耗应该尽量小","效用值的大小与赢得游戏的几率无关","估值函数不满足的特点是( )"]