知识卡片

普通读书笔记卡 · 6486.e · 生成模型:GPT-6

自编码以重建学习表示

理解等级:中级

自编码器把输入压缩或转换成内部表示,再学习重建输入。重建目标本身就是学习信号,不能说它“重建而不学习”。用于推荐时可以从评分或内容中提取特征,但重建好不自动等于排序好。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Auto Encoder (AE) is a special artificial neural network that uses its input to reconstruct data rather than learn. A sparse AE uses the high dimensional and scant representation to extract features from input data [35]. Deep AE is a kind of deep neural network and incorporates an unsupervised learning algorithm with the same input and output [36]. Deep AE is often pre-trained using a

    参考来源

    - 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第03章《深度推荐与评估指标》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C3.html) - 结论依据:摘录中关于自编码以重建学习表示的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Auto Encoder (AE) is a special artificial neural network that uses its input to reconstruct data rather than learn. A sparse AE uses the high dimensional and scant representation to extract features from input data [35]. Deep AE is a kind of deep neural network and incorporates an unsupervised learning algorithm with the same input and output [36]. Deep AE is often pre-trained using a