Convolutional Neural Networks (CNNs) models have been recently introduced in
the domain of top-$N$ session-based recommendations. An ordered collection of
past items the user has interacted with in a session (or sequence) are embedded
into a 2-dimensional latent matrix, and treated as an image. The convolution
and pooling operations are then applied to the mapped item embeddings. In this
paper, we first examine the typical session-based CNN recommender and show that
both the generative model and network architecture are suboptimal when modeling
long-range dependencies in the item sequence. To address the issues, we propose
a simple, but very effective generative model that is capable of learning
high-level representation from both short- and long-range dependencies. The
network architecture of the proposed model is formed of a stack of holed
convolutional layers, which can efficiently increase the receptive fields
without relying on the pooling operation. Another contribution is the effective
use of residual block structure in recommender systems, which can ease the
optimization for much deeper networks. The proposed generative model attains
state-of-the-art accuracy with less training time in the session-based
recommendation task. It accordingly can be used as a powerful session-based
recommendation baseline to beat in future, especially when there are long
sequences of user feedback.
Description
[1808.05163] A Simple but Hard-to-Beat Baseline for Session-based Recommendations
%0 Generic
%1 yuan2018simple
%A Yuan, Fajie
%A Karatzoglou, Alexandros
%A Arapakis, Ioannis
%A Jose, Joemon M
%A He, Xiangnan
%D 2018
%K recommendation session
%T A Simple but Hard-to-Beat Baseline for Session-based Recommendations
%U http://arxiv.org/abs/1808.05163
%X Convolutional Neural Networks (CNNs) models have been recently introduced in
the domain of top-$N$ session-based recommendations. An ordered collection of
past items the user has interacted with in a session (or sequence) are embedded
into a 2-dimensional latent matrix, and treated as an image. The convolution
and pooling operations are then applied to the mapped item embeddings. In this
paper, we first examine the typical session-based CNN recommender and show that
both the generative model and network architecture are suboptimal when modeling
long-range dependencies in the item sequence. To address the issues, we propose
a simple, but very effective generative model that is capable of learning
high-level representation from both short- and long-range dependencies. The
network architecture of the proposed model is formed of a stack of holed
convolutional layers, which can efficiently increase the receptive fields
without relying on the pooling operation. Another contribution is the effective
use of residual block structure in recommender systems, which can ease the
optimization for much deeper networks. The proposed generative model attains
state-of-the-art accuracy with less training time in the session-based
recommendation task. It accordingly can be used as a powerful session-based
recommendation baseline to beat in future, especially when there are long
sequences of user feedback.
@misc{yuan2018simple,
abstract = {Convolutional Neural Networks (CNNs) models have been recently introduced in
the domain of top-$N$ session-based recommendations. An ordered collection of
past items the user has interacted with in a session (or sequence) are embedded
into a 2-dimensional latent matrix, and treated as an image. The convolution
and pooling operations are then applied to the mapped item embeddings. In this
paper, we first examine the typical session-based CNN recommender and show that
both the generative model and network architecture are suboptimal when modeling
long-range dependencies in the item sequence. To address the issues, we propose
a simple, but very effective generative model that is capable of learning
high-level representation from both short- and long-range dependencies. The
network architecture of the proposed model is formed of a stack of holed
convolutional layers, which can efficiently increase the receptive fields
without relying on the pooling operation. Another contribution is the effective
use of residual block structure in recommender systems, which can ease the
optimization for much deeper networks. The proposed generative model attains
state-of-the-art accuracy with less training time in the session-based
recommendation task. It accordingly can be used as a powerful session-based
recommendation baseline to beat in future, especially when there are long
sequences of user feedback.},
added-at = {2018-09-16T20:50:12.000+0200},
author = {Yuan, Fajie and Karatzoglou, Alexandros and Arapakis, Ioannis and Jose, Joemon M and He, Xiangnan},
biburl = {https://www.bibsonomy.org/bibtex/2bf20932ef37086ccc0bf34b3fdd0f32f/e.fischer},
description = {[1808.05163] A Simple but Hard-to-Beat Baseline for Session-based Recommendations},
interhash = {4547816edb8618fef1ed6427dfd8e3aa},
intrahash = {bf20932ef37086ccc0bf34b3fdd0f32f},
keywords = {recommendation session},
note = {cite arxiv:1808.05163},
timestamp = {2018-09-16T20:50:12.000+0200},
title = {A Simple but Hard-to-Beat Baseline for Session-based Recommendations},
url = {http://arxiv.org/abs/1808.05163},
year = 2018
}