Hybrid Recommender System based on Autoencoders (2016-06-24T00:00:00.000000Z)

TL;DR

This paper enhanced the architecture of Recommender Systems by using a loss function adapted to input data with missing values, and by incorporating side information, demonstrating that while side information only slightly improve the test error averaged on all users/items, it has more impact on cold users/ items.

Abstract

A standard model for Recommender Systems is the Matrix Completion setting: given partially known matrix of ratings given by users (rows) to items (columns), infer the unknown ratings. In the last decades, few attempts where done to handle that objective with Neural Networks, but recently an architecture based on Autoencoders proved to be a promising approach. In current paper, we enhanced that architecture (i) by using a loss function adapted to input data with missing values, and (ii) by incorporating side information. The experiments demonstrate that while side information only slightly improve the test error averaged on all users/items, it has more impact on cold users/items.

Authors

Florian Strub

6 papers

Jérémie Mary

3 papers

R. Gaudel

1 papers

TL;DR

Abstract

Authors

References50 items

Recurrent Coevolutionary Feature Embedding Processes for Recommendation

Collaborative Recurrent Autoencoder: Recommend while Learning to Fill in the Blanks

Wide & Deep Learning for Recommender Systems

Low-Rank Matrix Approximation with Stability

A Network-based End-to-End Trainable Task-oriented Dialogue System

Collaborative Denoising Auto-Encoders for Top-N Recommender Systems

Deep Residual Learning for Image Recognition

Collaborative Filtering with Stacked Denoising AutoEncoders and Sparse Inputs

Neural Network Matrix Factorization

Representation Learning via Semi-Supervised Autoencoder for Multi-task Learning

Deep Collaborative Filtering via Marginalized Denoising Auto-encoder

WEMAREC: Accurate and Scalable Recommendation through Weighted and Ensemble Matrix Approximation

AutoRec: Autoencoders Meet Collaborative Filtering

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Improving Content-based and Hybrid Music Recommendation using Deep Learning

Collaborative Deep Learning for Recommender Systems

Scalable Variational Bayesian Matrix Factorization with Side Information

Social Popularity based SVD++ Recommender System

Deep content-based music recommendation

Distributed Representations of Words and Phrases and their Compositionality

Local Low-Rank Matrix Approximation

A Comparative Study of Collaborative Filtering Algorithms

Reconstructing Missing Data in State Estimation With Autoencoders

Deep Sparse Rectifier Neural Networks

Recommender systems with social regularization

Factorization Machines

Bayesian Matrix Factorization with Side Information and Dirichlet Process Mixtures

Understanding the difficulty of training deep feedforward neural networks

Incorporating Side Information in Probabilistic Matrix Factorization with Gaussian Processes

Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

Matrix Factorization Techniques for Recommender Systems

Ordinal Boltzmann Machines for Collaborative Filtering

Bayesian probabilistic matrix factorization using Markov chain Monte Carlo

Extracting and composing robust features with denoising autoencoders

Large-Scale Parallel Collaborative Filtering for the Netflix Prize

Probabilistic Matrix Factorization

Restricted Boltzmann machines for collaborative filtering

Efficient sparse coding algorithms

Training Neural Networks with Deficient Data

Nonlinear principal component analysis using autoassociative neural networks

Dirichlet Process Mixtures

Deep Learning

The netflix recommender system : Algorithms , business value , and innovation

Dropout: a simple way to prevent neural networks from overfitting

SVDFeature: a toolkit for feature-based collaborative filtering

Efficient BackProp

Content-based Recommender Systems: State of the Art and Trends

Active learning in regression, with application to stochastic dynamic programming

Neural Networks for Pattern Recognition

Field of Study

Journal Information

Name