AutoRec: Autoencoders Meet Collaborative Filtering

This paper proposes AutoRec, a novel autoencoder framework for collaborative filtering (CF). Empirically, AutoRec’s compact and efficiently trainable model outperforms stateof-the-art CF techniques (biased matrix factorization, RBMCF and LLORMA) on the Movielens and Netflix datasets.

PDF Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Recommendation Systems MovieLens 10M I-AutoRec RMSE 0.782 # 10
Recommendation Systems MovieLens 1M I-AutoRec RMSE 0.831 # 5

Methods