Abstract | ||
---|---|---|
Pairwise algorithms are popular for learning recommender systems from implicit feedback. For each user, or more generally context, they try to discriminate between a small set of selected items and the large set of remaining (irrelevant) items. Learning is typically based on stochastic gradient descent (SGD) with uniformly drawn pairs. In this work, we show that convergence of such SGD learning algorithms slows down considerably if the item popularity has a tailed distribution. We propose a non-uniform item sampler to overcome this problem. The proposed sampler is context-dependent and oversamples informative pairs to speed up convergence. An efficient implementation with constant amortized runtime costs is developed. Furthermore, it is shown how the proposed learning algorithm can be applied to a large class of recommender models. The properties of the new learning algorithm are studied empirically on two real-world recommender system problems. The experiments indicate that the proposed adaptive sampler improves the state-of-the art learning algorithm largely in convergence without negative effects on prediction quality or iteration runtime. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1145/2556195.2556248 | WSDM |
Keywords | Field | DocType |
proposed adaptive sampler,real-world recommender system problem,proposed sampler,recommender system,pairwise algorithm,improving pairwise,implicit feedback,item recommendation,item popularity,constant amortized runtime cost,recommender model,new learning algorithm,non-uniform item sampler,matrix factorization,recommender systems | Convergence (routing),Data mining,Computer science,Popularity,Theoretical computer science,Artificial intelligence,Small set,Speedup,Recommender system,Pairwise comparison,Stochastic gradient descent,Information retrieval,Matrix decomposition,Machine learning | Conference |
Citations | PageRank | References |
113 | 2.83 | 21 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Steffen Rendle | 1 | 1963 | 70.68 |
Christoph Freudenthaler | 2 | 1853 | 61.55 |