Scalable Recommender Systems through Recursive Evidence Chains

TL;DR

This paper introduces a scalable recommender system that generates latent variables on demand using evidence chains, effectively addressing cold-start and online learning challenges while maintaining competitive accuracy and speed.

Contribution

It proposes a novel recursive evidence chain method for matrix completion that scales efficiently and handles cold-start and online learning scenarios.

Findings

Achieves competitive accuracy compared to existing matrix factorization methods.

Demonstrates improved convergence speed and scalability.

Effectively addresses cold-start and online learning problems.

Abstract

Recommender systems can be formulated as a matrix completion problem, predicting ratings from user and item parameter vectors. Optimizing these parameters by subsampling data becomes difficult as the number of users and items grows. We develop a novel approach to generate all latent variables on demand from the ratings matrix itself and a fixed pool of parameters. We estimate missing ratings using chains of evidence that link them to a small set of prototypical users and items. Our model automatically addresses the cold-start and online learning problems by combining information across both users and items. We investigate the scaling behavior of this model, and demonstrate competitive results with respect to current matrix factorization techniques in terms of accuracy and convergence speed.