Filters








16,160 Hits in 5.6 sec

Modeling Sequences as Distributions with Uncertainty for Sequential Recommendation [article]

Ziwei Fan, Zhiwei Liu, Lei Zheng, Shen Wang, Philip S. Yu
2021 arXiv   pre-print
In this work, we propose a Distribution-based Transformer for Sequential Recommendation (DT4SR), which injects uncertainties into sequential modeling.  ...  We devise two novel Trans-formers for modeling mean and covariance, which guarantees the positive-definite property of distributions.  ...  PROBLEM DEFINITION A recommender system collects feedback between a set of users U and items V, e.g., clicks.  ... 
arXiv:2106.06165v1 fatcat:l3yosbzu3bbdtbhspmas5umvuy

Distributional Contrastive Embedding for Clarification-based Conversational Critiquing

Tianshu Shen, Zheda Mai, Ga Wu, Scott Sanner
2022 Proceedings of the ACM Web Conference 2022  
embedding-based recommendation methods.  ...  In summary, this work adds a new dimension of clarification to enhance the well-known critiquing framework along with a novel data-driven distributional embedding for clarification suggestions that significantly  ...  CONCLUSION We presented a novel methodology for clarification in critiquingbased conversational recommender systems.  ... 
doi:10.1145/3485447.3512114 fatcat:honbjovihrfttfvaoyviox6tyi

Convolutional Gaussian Embeddings for Personalized Recommendation with Uncertainty [article]

Junyang Jiang and Deqing Yang and Yanghua Xiao and Chenlu Shen
2020 arXiv   pre-print
Most of existing embedding based recommendation models use embeddings (vectors) corresponding to a single fixed point in low-dimensional space, to represent users and items.  ...  Such embeddings fail to precisely represent the users/items with uncertainty often observed in recommender systems.  ...  For example, the authors in [Sedhain et al., 2015] proposed a novel AutoEncoder (AE) framework for CF. DeepMF [Xue et al., 2017] is a deep version of MF-based recommendation model.  ... 
arXiv:2006.10932v1 fatcat:4wjxcknbbbgdhhbgtzwawx73ny

Collaborative Filtering Based on a Variational Gaussian Mixture Model

FengLei Yang, Fei Liu, ShanShan Liu
2021 Future Internet  
This paper proposes a variational autoencoder that uses a Gaussian mixture model for latent factors distribution for CF, GVAE-CF.  ...  Collaborative filtering (CF) is a widely used method in recommendation systems.  ...  [16] introduced a recommender VAE (RecVAE) model that uses a novel composite prior distribution for the latent representation.  ... 
doi:10.3390/fi13020037 fatcat:y3exk6ifcrfhzcxr5ogkvuch24

Exploiting Variational Domain-Invariant User Embedding for Partially Overlapped Cross Domain Recommendation [article]

Weiming Liu, Xiaolin Zheng, Mengling Hu, Chaochao Chen
2022 arXiv   pre-print
To address this issue, we propose an end-to-end dual-autoencoder with Variational Domain-invariant Embedding Alignment (VDEA) model, a cross-domain recommendation framework for the POCDR problem, which  ...  Cross-Domain Recommendation (CDR) has been popularly studied to utilize different domain knowledge to solve the cold-start problem in recommender systems.  ...  INTRODUCTION Recommender systems become more and more indispensable in practice since it is a sharp tool for users to easily find what they desire.  ... 
arXiv:2205.06440v1 fatcat:kbxzmkt3pndkrpentjvuk564xi

Convolutional-NN and Word Embedding for Making an Effective Product Recommendation Based on Enhanced Contextual Understanding of a Product Review

Hanafi, Nanna Suryana, Abd Samad Bin Hasan Basari
2019 International Journal on Advanced Science, Engineering and Information Technology  
Collaborative filtering is a model of a recommendation algorithm that relies on rating as the fundamental calculation to make a recommendation. It has been successfully implemented in e-commerce.  ...  Our method outperformed several previous works based on a metric evaluation using the Root Mean Squared Error (RMSE).  ...  ACKNOWLEDGMENT We would like to thank for some institution that is often supporting us to conduct research that are Universitas Amikom, Universiti Teknikal Malaysia Melaka, PT. Time Excelindo.  ... 
doi:10.18517/ijaseit.9.3.8843 fatcat:tn6oh66xnzg6xfpa36ogvpa5ae

Sequential Recommendation via Stochastic Self-Attention [article]

Ziwei Fan, Zhiwei Liu, Alice Wang, Zahra Nazari, Lei Zheng, Hao Peng, Philip S. Yu
2022 arXiv   pre-print
Sequential recommendation models the dynamics of a user's previous behaviors in order to forecast the next item, and has drawn a lot of attention.  ...  We devise a novel Wasserstein Self-Attention module to characterize item-item position-wise relationships in sequences, which effectively incorporates uncertainty into model training.  ...  Specifically, we model items as Gaussian distributions with stochastic embeddings, consisting of mean (for base interests) and covariance (for the variability of interests) embeddings.  ... 
arXiv:2201.06035v2 fatcat:h27uaasfzjf2ngb4sy2d36rs5e

Xbox movies recommendations

Noam Koenigstein, Ulrich Paquet
2013 Proceedings of the 7th ACM conference on Recommender systems - RecSys '13  
The model is designed for binary feedback, which is common in many real-world systems where numeric rating data is scarce or non-existent.  ...  We present a matrix factorization model inspired by challenges we encountered while working on the Xbox movies recommendation system.  ...  CONCLUSION We presented a novel probabilistic Matrix Factorization model with Embedded Feature Selection (MF-EFS) for Xbox Movies.  ... 
doi:10.1145/2507157.2507168 dblp:conf/recsys/KoenigsteinP13 fatcat:2uqipsiiove5bknrr6dqo3zvn4

Bayesian Embeddings for Few-Shot Open World Recognition [article]

John Willes, James Harrison, Ali Harakeh, Chelsea Finn, Marco Pavone, Steven Waslander
2021 arXiv   pre-print
We combine Bayesian non-parametric class priors with an embedding-based pre-training scheme to yield a highly flexible framework which we refer to as few-shot learning for open world recognition (FLOWR  ...  This stands in stark contrast to modern machine learning systems that are typically designed with a known set of classes and a large number of examples for each class.  ...  We turn to a non-parametric embedding-based few-shot learning framework based on the Dirichlet process Gaussian mixture model [11] , [32] .  ... 
arXiv:2107.13682v1 fatcat:odbdstuw2vavvhuue5bwoj2kae

Graph Embedding for Recommendation against Attribute Inference Attacks [article]

Shijie Zhang, Hongzhi Yin, Tong Chen, Zi Huang, Lizhen Cui, Xiangliang Zhang
2021 arXiv   pre-print
In recent years, recommender systems play a pivotal role in helping users identify the most suitable items that satisfy personal preferences.  ...  In our paper, we propose GERAI, a novel differentially private graph convolutional network to address such limitations.  ...  In this paper, we subsume the GCN-based recommender under the differential privacy (DP) constraint, and propose a novel privacy-preserving recommender GERAI, namely Graph Embedding for Recommendation against  ... 
arXiv:2101.12549v1 fatcat:mae3fmklcjc3dk32y6uyyvju7q

Variational Bandwidth Auto-encoder for Hybrid Recommender Systems [article]

Yaochen Zhu, Zhenzhong Chen
2021 arXiv   pre-print
In this article, we propose a variational bandwidth auto-encoder (VBAE) for recommendations, aiming to address the sparsity and noise problems simultaneously.  ...  VBAE first encodes user collaborative and feature information into Gaussian latent variables via deep neural networks to capture non-linear user similarities.  ...  sources, i.e., ratings and user features in recommender systems.  ... 
arXiv:2105.07597v2 fatcat:eajcoprjefeorbr4qsklmwerwa

Graph Embedding for Recommendation against Attribute Inference Attacks

Shijie Zhang, Hongzhi Yin, Tong Chen, Zi Huang, Lizhen Cui, Xiangliang Zhang
2021 Proceedings of the Web Conference 2021  
is the non-private, GCN-based recommendation model proposed in [55] . • Blurm: This method directly uses perturbed user-item ratings to train the recommender system [50] . • DPAE: In DPAE, Gaussian  ...  In this paper, we subsume the GCN-based recommender under the differential privacy (DP) constraint, and propose a novel privacy-preserving recommender GERAI, namely Graph Embedding for Recommendation against  ... 
doi:10.1145/3442381.3449813 fatcat:sbdyykzmrfho3c36g7o6gy6o6y

Group-sparse Embeddings in Collective Matrix Factorization [article]

Arto Klami, Guillaume Bouchard, Abhishek Tripathi
2014 arXiv   pre-print
A typical example is the joint modeling of user-item, item-property, and user-feature matrices in a recommender system.  ...  CMF is a technique for simultaneously learning low-rank representations based on a collection of matrices with shared entities.  ...  Non-Gaussian observations For non-Gaussian data we use the approximation schema presented by Seeger and Bouchard (2012) , adaptively approximating non-Gaussian likelihoods with spherical-variance Gaussians  ... 
arXiv:1312.5921v2 fatcat:7rzcpupa3jbuvmju4st34q7sd4

Light Graph Convolutional Collaborative Filtering with Multi-aspect Information

Denghua Mei, Niu Huang, Xin Li
2021 IEEE Access  
In this paper, we propose a Light GCN based Aspect-level Collaborative Filtering model (LGC-ACF) to exploit multi-aspect user-item interaction information.  ...  INDEX TERMS Recommender systems, graph convolutional network, representation learning, multi-aspect information.  ...  For our proposed model, we initialize the model parameters with a Gaussian distribution of mean 0 and standard deviation 0.1, and apply grid research for learning the hyperparameters.  ... 
doi:10.1109/access.2021.3061915 fatcat:yeus775p5nebvcrpjbzqse7f7q

Style conditioned recommendations

Murium Iqbal, Kamelia Aryafar, Timothy Anderton
2019 Proceedings of the 13th ACM Conference on Recommender Systems - RecSys '19  
We show a 12% improvement on NDCG@20 over the traditional VAE based approach and an average 22% improvement on AUC across all classes for predicting user style profiles against our best performing baseline  ...  We propose Style Conditioned Recommendations (SCR) and introduce style injection as a method to diversify recommendations.  ...  We chose to compare to a Random Forest classifier as well, as it is a non-linear ensemble model. We allowed for 1000 trees in the Random Forest and no maximum depth defined.  ... 
doi:10.1145/3298689.3347007 dblp:conf/recsys/IqbalAA19 fatcat:wcngwunznva7pmbyr6dlccwyd4
« Previous Showing results 1 — 15 out of 16,160 results