Probabilistic Matrix Factorization

By Ruslan Salakhutdinov, Andriy Mnih. 2007

4,825 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods, Recommender Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

In Recommender Systems

Citations in the corpus

Collaborative Deep Learning for Recommender Systems (2014, methodology): “It is a probabilistic graphical model that seamlessly integrates a topic model, latent Dirichlet allocation (LDA) [5], and a model-based CF method, probabilistic matrix factorization (PMF) [27].”
Large scale visual recommendations from street fashion images (2014, methodology): “We also attempted adopting the collaborative filtering formulation of [11], but do not describe it in detail due to practical difficulties we encountered in scaling the algorithm to large data sets.”
Neural Collaborative Filtering (2017, background): “ractions; and w ui is a hyperparameter denoting the weight of training instance (u;i). While the squared loss can be explained by assuming that observations are generated from a Gaussian distribution [29], we point out that it may not tally well with implicit data. This is because for implicit data,”
A Neural Autoregressive Approach to Collaborative Filtering (2016, methodology): “A prominent example is RBM-based CF (RBM-CF) (Salakhutdinov et al., 2007).”
Learning Distributed Representations from Reviews for Collaborative Filtering (2015, background): “In this setting, collaborative filtering is usually cast as a problem of matrix factorization with missing values [10, 16, 18].”
Explainable Recommendation: A Survey and New Perspectives (2018, background): “For example, Mobasher et al. [2001] leveraged association rule mining for efficient web page recommendation at large-scale; Cho et al.”
How algorithmic confounding in recommendation systems increases homogeneity and decreases utility (2017, background): “Probabilistic matrix factorization [49] assumes that each cell in the user preference and item attribute matrices are normally distributed, whereas a probabilistic construc-”
Word Representations via Gaussian Embedding (2014, background): “Related work in probabilistic matrix factorization (Mnih & Salakhutdinov, 2007) embeds rows and columns as Gaussians, and some forms of this do provide each row and column with its own variance (Salakhutdinov & Mnih, 2008).”
Ask the GRU: Multi-task Learning for Deep Text Recommendations (2016, methodology): “Classical examples of this include matrix factorization methods [6, 7] which completely ignore text information and rely solely on the rating matrix.”
Freeze-Thaw Bayesian Optimization (2014, methodology): “Probabilistic Matrix Factorization As a ﬁnal experiment, we optimize three hyperparameters of a probabilistic matrix factorization (PMF) [24] on 100,000 ratings from the MovieLens data set [25].”
Joint Deep Modeling of Users and Items Using Reviews for Recommendation (2017, methodology): “techniques. •It does not only alleviate the problem of sparsity by leveraging reviews, but also improves the overall performance of the system signiﬁcantly. It outperforms state-of-the-art techniques [17, 33, 25, 35] in terms of prediction accuracy on all of the evaluated datasets including Yelp, 21”
SoRec: social recommendation using probabilistic matrix factorization (2008, methodology): “4.3 Comparison In this section, in order to show the performance improvement of our Social Recommendation (SoRec) algorithm, we compare our algorithm with some state-of-the-art algorithms: Maximum Margin Matrix Factorization (MMMF) [15], Probabilistic Matrix Factorization (PMF) [18], and Constr”
Deep Exponential Families (2014, methodology): “We compare double DEFs to l2-regularized (Gaussian) matrix factorization (MF) [32].”
Exploiting compositionality to explore a large space of model structures (2012, methodology): “low-rank approximation (Salakhutdinov and Mnih, 2008) Bayesian clustered tensor factorization (Sutskever et al.”