FM

今天分享的是2010年由Steffen Rendle于2010年在ICDM上提出的推荐系统论文《Factorization Machines》

#一、摘要

FM（Factorization Machines）模型优势如下：

1. FM在数据量比较大并且特征稀疏的情况下，仍然能够得到优秀的性能。
2. FM可以在线性时间对新样本作出预测。
3. FM可以处理任何实值特征向量。（？？？？？哈 这是为什么啊？看不明白FMs are a general predictor that can work with any real valued feature vector. In contrast to this, other state-of- the-art factorization models work only on very restricted input data.）

是用户矩阵，

#待完善

1. feature-base 有交互特征《深度学习中的匹配》
2. 那一大推fm的问题。 ali
3. 模型的表达能力指的是？
4. MF 等模型没有泛化性？？是针对特定的数据提出的？On the other hand, the drawback of tensor factorization models and even more for specialized factorization models is that (1) they are not applicable to standard prediction data (e.g. a real valued feature vector in R .) and (2) that specialized models are usually derived individually for a specific task requiring effort in modelling and design of a learning algorithm.On the other hand there are many different factorization mod- els like matrix factorization, parallel factor analysis or specialized models like SVD++, PITF or FPMC. The drawback of these models is that they are not applicable for general prediction tasks but work only with special input data。Furthermore their model equations and optimization algorithms are derived individually for each task. 5.

#参考链接
[1] 扒一扒 FM 算法的实现
[2] Deep FM算法代码级精讲
[3] FM因子分解机模型的原理、推导、代码和应用
[4] FM git
[5] FM算法复现