[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fIrFLClhlpTein5PPC3TMEEtOAJVN5w93HtXGK1aun4c":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":19,"related":20,"source":30,"type":31},[],"2024-01-06 15:03:29",127969806,[8,9,10,11],"一种机器学习方法","通过多层神经网络学习复杂的特征和表示","提取和理解数据的高层次抽象特征","实现对大规模数据集的高效学习和预测",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":18},20,"ec28b52071c9a494c034f273cf435853","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fa52ccb88374a2afd2852f71ccdfdd428.png","人工智能基础及应用","337bc9d59c8e484da140b1590cac6c1e","人工智能技术作业","深度学习是",[21,32,41,44,53,63,72,81,90,95],{"answer":22,"createTime":5,"id":23,"options":24,"question":29,"source":30,"type":31},[],127969804,[25,26,27,28],"输入层","神经层","隐藏层","输出层","神经网络由大量互相连接的神经元组成,形成层次结构,包括哪三层","v1",1,{"answer":33,"createTime":5,"id":34,"options":35,"question":40,"source":30,"type":31},[],127969805,[36,37,38,39],"算法通过调整模型参数,从数据输出中学习并具备自学与改进的能力","不需要硬性规则,而是通过观察可访问的数据集并比较实际输出来学习","发现数据中的模式,通过调整模型以最小化预测输出与实际输出的差异","简单地逆向工程产生输出","机器学习是指",{"answer":42,"createTime":5,"id":6,"options":43,"question":19,"source":30,"type":31},[],[8,9,10,11],{"answer":45,"createTime":5,"id":46,"options":47,"question":52,"source":30,"type":31},[],127969807,[48,49,50,51],"通过深度神经网络模型进行端到端的特征学习和模式识别","机器学习","逻辑推理","超级算力","深度学习在图像识别、语音处理、自然语言处理等领域取得显著成功,其核心是",{"answer":54,"createTime":55,"id":56,"options":57,"question":62,"source":30,"type":31},[],"2024-01-06 15:03:30",127969808,[58,59,60,61],"图像识别","语音处理","自然语言处理","金融预测","深度学习可用于哪些领域",{"answer":64,"createTime":55,"id":65,"options":66,"question":71,"source":30,"type":31},[],127969809,[67,68,69,70],"通过深度学习算法,计算机能够从文本数据中提取特征","实现对自然语言的理解","以自然语言生成输出结果","使计算机与人类之间的沟通与理解变得更为顺畅,再无语言障碍","自然语言处理是",{"answer":73,"createTime":55,"id":74,"options":75,"question":80,"source":30,"type":31},[],127969810,[76,77,78,79],"大型语言模型在很大程度上代表了一类称为Transformer深度学习架构","它是一种神经网络","2017年由Google研发团队在"Attention is All You Need"论文中首次提出,Transformer模型的设计基于自注意力机制(Self-Attention),这是一种可以在输入序列的不同位置关注不同部分的机制","一种处理世界上最庞大语言的模型","什么是大语言模型呢",{"answer":82,"createTime":83,"id":84,"options":85,"question":88,"source":30,"type":89},[],"2024-12-05 22:31:41",169240574,[86,87],"正确","错误","神经网络是一种受到生物神经系统启发而设计的计算模型,由大量互相连接的神经元组成",3,{"answer":91,"createTime":83,"id":92,"options":93,"question":94,"source":30,"type":89},[],169240575,[86,87],"神经网络的神经元通过具有调整权重的连接传递信息,而激活函数决定神经元的输出",{"answer":96,"createTime":83,"id":97,"options":98,"question":99,"source":30,"type":89},[],169240576,[86,87],"神经网络在训练过程中,通过前向传播将数据从输入传递到输出,然后通过反向传播根据损失函数的梯度调整连接权重,优化网络性能"]