Latent Factor Model Fitting Algorithm

A Latent Factor Model Fitting Algorithm is a statistical modeling algorithm that fits a latent factor model (from some latent factor model family).

• Context:
  • It can be implemented by a Latent Factor Model Fitting System (that solves a latent factor modeling task).
  • …
• Example(s):
  • Matrix Factorization-based Learning Algorithm, such as a matrix factorization-based item recommendations algorithm.
  • Reconstruction Algorithm (Nickel et al., 2011)
  • Clustering Algorithm (Kok & Domingos, 2007),
  • Ranking Costs Algorithm.
  • a Latent Factor Item Recommendation Algorithm.
  • …
• Counter-Example(s):
  • Decision Tree-based Algorithm.
• See: Latent Vector.

---

---

References

2017

• (Zhao et al., 2017) ⇒ Qian Zhao, Yue Shi, and Liangjie Hong. (2017). “GB-CENT: Gradient Boosted Categorical Embedding and Numerical Trees.” In: Proceedings of the 26th International Conference on World Wide Web. ISBN:978-1-4503-4913-0 doi:10.1145/3038912.3052668
  • QUOTE: Latent factor models and decision tree based models are widely used in tasks of prediction, ranking and recommendation. Latent factor models have the advantage of interpreting categorical features by a low-dimensional representation, while such an interpretation does not naturally fit numerical features. In contrast, decision tree based models enjoy the advantage of capturing the nonlinear interactions of numerical features, while their capability of handling categorical features is limited by the cardinality of those features. …

2011

• (Wang & Blei, 2011) ⇒ Chong Wang, and David M. Blei. (2011). “Collaborative Topic Modeling for Recommending Scientific Articles.” In: Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ISBN:978-1-4503-0813-7 doi:10.1145/2020408.2020480
  • QUOTE: Among latent factor methods, matrix factorization performs well [13]. In matrix factorization, we represent users and items in a shared latent low-dimensional space of dimension [math]\displaystyle{ K }[/math] — user [math]\displaystyle{ i }[/math] is represented by a latent vector [math]\displaystyle{ u_j \in \mathbb{R}^K }[/math] and item [math]\displaystyle{ j }[/math] by a latent vector [math]\displaystyle{ v_j \in \mathbb{R}^K }[/math]. We form the prediction of whether user [math]\displaystyle{ i }[/math] will like item [math]\displaystyle{ j }[/math] with the inner product between their latent representations, [math]\displaystyle{ \hat{r}_{ij} = u^T_iv_j. \ (1) }[/math]

2009

• (Agarwal et al., 2009) ⇒ Deepak Agarwal, and Bee-Chung Chen. (2009). “Regression-based Latent Factor Models.” In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2009). doi:10.1145/1557019.1557029

---