[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$flhQs7WuLfgAL2lW7fL0A4WVbcX92GzM1VtnN4-Ow51w":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":19,"related":20,"source":31,"type":32},[],"2025-05-06 18:29:53",185941554,[8,9,10,11],"标签信息","真实样本","潜在空间向量","随机图像",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":18},31,"034a2b5846d1eb1c9da37ec2b6f32f52","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F5db85478aa0dbac2d08489d9652f8926.png","智能图像处理","5ef1eba13df1498f8023cbfaaff21d6d","","在生成对抗网络(GAN)中,生成器通过______生成图像.",[21,33,43,52,61,70,79,84,93,102],{"answer":22,"createTime":23,"id":24,"options":25,"question":30,"source":31,"type":32},[],"2025-05-06 18:26:40",185941343,[26,27,28,29],"区分真假样本","生成与真实样本分布一致的样本","训练判别模型","提高判别模型的性能","在生成对抗网络(GAN)中,生成模型的目的是______","v1",0,{"answer":34,"createTime":35,"id":36,"options":37,"question":42,"source":31,"type":32},[],"2025-05-06 18:26:41",185941344,[38,39,40,41],"生成样本","将生成样本与真实样本区分开","增加生成模型的训练速度","学习生成样本的特征","在GAN的训练过程中,判别模型的任务是______",{"answer":44,"createTime":35,"id":45,"options":46,"question":51,"source":31,"type":32},[],185941345,[47,48,49,50],"卷积神经网络(CNN)","转置卷积(Deconvolution)","全连接层","LSTM网络","深度卷积生成对抗神经网络(DCGAN)的生成模型使用______来生成图像",{"answer":53,"createTime":35,"id":54,"options":55,"question":60,"source":31,"type":32},[],185941346,[56,57,58,59],"随机噪声","标签或条件约束","卷积层","深度学习优化算法","条件生成对抗网络(CGAN)的主要改进是引入了______",{"answer":62,"createTime":35,"id":63,"options":64,"question":69,"source":31,"type":32},[],185941347,[65,66,67,68],"图像分类","图像生成","无配对样本的图像到图像的转换","序列数据生成","CycleGAN主要用于______",{"answer":71,"createTime":35,"id":72,"options":73,"question":78,"source":31,"type":32},[],185941348,[74,75,76,77],"对抗损失","循环一致损失","恒等变换损失","以上所有","在CycleGAN中,______用于确保图像转换的效果",{"answer":80,"createTime":35,"id":81,"options":82,"question":83,"source":31,"type":32},[],185941349,[10,9,11,8],"在生成对抗网络(GAN)中,生成器通过______生成图像",{"answer":85,"createTime":35,"id":86,"options":87,"question":92,"source":31,"type":32},[],185941350,[88,89,90,91],"同时更新","交替更新","随机更新","不更新","在训练GAN时,判别器和生成器是通过______进行优化的",{"answer":94,"createTime":35,"id":95,"options":96,"question":101,"source":31,"type":32},[],185941351,[97,98,99,100],"生成器与判别器之间的对抗","训练过程中的随机性","网络层之间的竞争","不同优化算法之间的竞争","生成对抗网络(GAN)的训练过程中的"博弈"指的是______",{"answer":103,"createTime":35,"id":104,"options":105,"question":110,"source":31,"type":32},[],185941352,[106,107,108,109],"生成样本和条件标签","条件标签和输入噪声","真实样本和条件标签","条件标签和目标图像","在CGAN中,生成模型的输入包括______"]