Bayesian Probabilistic Matrix Factorization: A User Frequency Analysis
Matrix factorization (MF) has become a common approach to collaborative filtering, due to ease of implementation and scalability to large data sets. Two existing drawbacks of the basic model is that it does not incorporate side information on either users or items, and assumes a common variance for all users. We extend the work of constrained probabilistic matrix factorization by deriving the Gibbs updates for the side feature vectors for items (Salakhutdinov and Minh, 2008). We show that this Bayesian treatment to the constrained PMF model outperforms simple MAP estimation. We also consider extensions to heteroskedastic precision introduced in the literature (Lakshminarayanan, Bouchard, and Archambeau, 2011). We show that this tends result in overfitting for deterministic approximation algorithms (ex: Variational inference) when the observed entries in the user / item matrix are distributed in an non-uniform manner. In light of this, we propose a truncated precision model. Our experimental results suggest that this model tends to delay overfitting.
Code (0)
등록된 구현이 없습니다.
Tasks
Collaborative FilteringVariational InferenceSimilar Papers 제목 키워드 기반
Bayesian Probabilistic Matrix Factorization
Matrix factorization is a widely used technique in recommendation systems. Probabilistic Matrix Factorization (PMF) [1] extends traditional matrix factorization by incorporating probability distributions over latent fact…
Bayesian InferenceComputational EfficiencyRecommendation SystemsUncertainty Quantification+1Kernelized Bayesian Matrix Factorization
We extend kernelized matrix factorization with a fully Bayesian treatment and with an ability to work with multiple side information sources expressed as different kernels. Kernel functions have been introduced to matrix…
Recommendation SystemsLinearly constrained Bayesian matrix factorization for blind source separation
We present a general Bayesian approach to probabilistic matrix factorization subject to linear constraints. The approach is based on a Gaussian observation model and Gaussian priors with bilinear equality and inequality …
blind source separationPrior Specification for Bayesian Matrix Factorization via Prior Predictive Matching
The behavior of many Bayesian models used in machine learning critically depends on the choice of prior distributions, controlled by some hyperparameters that are typically selected by Bayesian optimization or cross-vali…
Bayesian OptimizationHyperparameter OptimizationStochastic OptimizationA Bayesian Perspective for Determinant Minimization Based Robust Structured Matrix Factorizatio
We introduce a Bayesian perspective for the structured matrix factorization problem. The proposed framework provides a probabilistic interpretation for existing geometric methods based on determinant minimization. We mod…