[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fQGPpmNbSmCodMGihA98kV-5EUOysGCSNDVrJwH4CAB8":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":8,"question":15,"related":16,"source":27,"type":60},[],"2024-12-03 21:00:46",168812248,[],{"count":9,"courseId":10,"courseImg":11,"courseName":12,"workId":13,"workName":14},5,"546e42810d355f1f9b27c49e13af87f9","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F25b30343053994e8940089572d36015b.jpg","人工智能概论通识课（AI2001）","0c5b408510e340bba178bfd285ac2141","第二讲章节测试","在神经网络训练过程中,反向传播算法用于计算损失函数相对于网络参数的梯度,并通过这些梯度来更新网络的权重和偏置.这个过程从____开始,逐层向输入层传播梯度",[17,29,38,47,57],{"answer":18,"createTime":19,"id":20,"options":21,"question":26,"source":27,"type":28},[],"2024-12-03 20:43:06",168806421,[22,23,24,25],"反向误差传播是一种随机优化算法","反向误差传播不依赖于网络中的激活函数","反向误差传播利用链式法则来计算每层权重的误差梯度","反向误差传播只能用于单层网络的训练","以下关于反向误差传播(Backpropagation)的描述哪项是正确的","v1",0,{"answer":30,"createTime":19,"id":31,"options":32,"question":37,"source":27,"type":28},[],168806422,[33,34,35,36],"单层感知机可以解决任何分类问题","单层感知机只能处理线性可分的问题","单层感知机包含至少一个隐藏层","单层感知机不需要激活函数","单层感知机是一种简单的神经网络模型,以下关于单层感知机的描述哪一个是正确的",{"answer":39,"createTime":19,"id":40,"options":41,"question":46,"source":27,"type":28},[],168806423,[42,43,44,45],"深度学习是一种独立的算法,与机器学习无关",""深度"这个形容词在深度学习中指的是网络的输入层和输出层之间的距离","线性感知器可以成为通用分类器,不需要非线性激活函数","深度学习使用的是具有多层结构的神经网络来进行数据的表征学习","关于深度学习的描述,以下哪项是正确的",{"answer":48,"createTime":19,"id":49,"options":50,"question":55,"source":27,"type":56},[],168806424,[51,52,53,54],"梯度下降法是一种用于优化神经网络权重的算法","在梯度下降法中,学习率(learning rate)决定了权重更新的幅度","梯度下降法的目的是最小化损失函数,使模型预测更准确","在梯度下降法中,如果学习率设置得太高,可能会导致算法越过最小值点,从而无法收敛","关于神经网络中的梯度下降法,以下哪些描述是正确的",1,{"answer":58,"createTime":5,"id":6,"options":59,"question":15,"source":27,"type":60},[],[],2]