知识卡片
深层表示增加适配与调参
深度推荐可从文本、图像、交互和序列学习多层表示,表达非线性关系;同时增加数据需求、调参工作与解释困难。表示能力更强不证明必然优于简单模型,应把额外开销与目标数据上的收益一起比较。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
deep learning approaches have limitations like interpretability, data requirement, and extensive hyperparameter tuning [52]. Deep learning-based RS are classified as integration and neural network models. The integration model combines traditional systems with deep learning and deep learning systems. The neural network models have a single technique and a deep composite model. A deep learning framework uses historical data for training. During this process, new information
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:摘录中关于深层表示增加适配与调参的定义、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:deep learning approaches have limitations like interpretability, data requirement, and extensive hyperparameter tuning [52]. Deep learning-based RS are classified as integration and neural network models. The integration model combines traditional systems with deep learning and deep learning systems. The neural network models have a single technique and a deep composite model. A deep learning framework uses historical data for training. During this process, new information