Advertisement

WITS 2020 pp 179-188 | Cite as

Variational Autoencoders Versus Denoising Autoencoders for Recommendations

Conference paper
  • 14 Downloads
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 745)

Abstract

Recommender systems help users explore new content such as music and news by showing them what they will find potentially interesting. There are many methods and algorithms that can help recommender systems create personalized recommendations. All recommendation approaches can be divided into three categories: Content-based recommendation, Collaborative filtering and Hybrid methods. In this paper, we explore and compare Variational Autoencoders and Denoising Autoencoders for Collaborative Filtering with implicit feedback. A Variational Autoencoders(VAE) is a non-linear model, so it can capture patterns that are more complex in the data and since the forward pass is sufficient to obtain the recommendation of a given user then the query time is fast. A Denoising AutoEncoder is a specific type of AutoEncoder, which is generally classed as a type of deep neural network and is trained to use a hidden layer to reconstruct a particular model based on its inputs. Comparison results between Variational Autoencoders (VAE) and Denoising Autoencoders (DAE) show that VAE has the upper hand when it comes to large datasets while DAE is better when using small datasets. We explore and evaluate both methods on three public datasets and using different metrics.

Keywords

Denoising AutoEncoder (DAE) Variational AutoEncoder (VAE) Collaborative filtering Recommender systems 

References

  1. 1.
    Bobadilla J, Ortega F, Hernando A, Gutiérrez A (2013) Recommender systems survey. Knowl Based Syst 46:109–132.  https://doi.org/10.1016/j.knosys.2013.03.012CrossRefGoogle Scholar
  2. 2.
    Batmaz Z, Yurekli A, Bilge A, Kaleli C (2018) A review on deep learning for recommender systems: challenges and remedies. Artif Intell Rev.  https://doi.org/10.1007/s10462-018-9654-yCrossRefGoogle Scholar
  3. 3.
    Hossein A, Rafsanjani N, Salim N, Aghdam AR (2013) Recommendation systems: a review Karamollah Bagheri Fard. Int J Comput Eng Res 3(5):47–52Google Scholar
  4. 4.
    Nilashi M, Bagherifard K, Ibrahim O, Alizadeh H, Nojeem LA, Roozegar N (2013) Collaborative filtering recommender systems. Res J Appl Sci Eng Technol 5(16):4168–4182.  https://doi.org/10.1007/978-3-540-72079-9_9CrossRefGoogle Scholar
  5. 5.
    Miao Y, Yu L, Blunsom P (2016) Neural variational inference for text processing. In: 33rd international conference on machine learning, ICML 2016, vol 4, no. Mcmc, pp 2589–2600Google Scholar
  6. 6.
    Wang H, Wang N, Yeung DY (2015) Collaborative deep learning for recommender systems. In: Proceedings of ACM SIGKDD international conference on knowledge discovery and data mining, vol 2015, pp 1235–1244,  https://doi.org/10.1145/2783258.2783273
  7. 7.
    Liang D, Krishnan RG, Hoffman MD, Jebara T (2018) Variational autoencoders for collaborative filtering. In: Web conference on 2018—proceedings of world wide web conference WWW 2018, pp 689–698.  https://doi.org/10.1145/3178876.3186150
  8. 8.
    Doersch C (2016) Tutorial on variational autoencoders, pp 1–23 [Online]. Available: https://arxiv.org/abs/1606.05908
  9. 9.
    Xu L, Cao M, Song B, Zhang J, Liu Y, Alsaadi FE (2018) Open-circuit fault diagnosis of power rectifier using sparse autoencoder based deep neural network. Neurocomputing 311:1–10.  https://doi.org/10.1016/j.neucom.2018.05.040CrossRefGoogle Scholar
  10. 10.
    Dolz J et al (2016) Stacking denoising auto-encoders in a deep network to segment the brainstem on MRI in brain cancer patients: a clinical study. Comput Med Imaging Graph 52:8–18.  https://doi.org/10.1016/j.compmedimag.2016.03.003CrossRefGoogle Scholar
  11. 11.
    Su X, Khoshgoftaar TM (2009) A survey of collaborative filtering techniques. Adv Artif Intell 2009(3):1–19.  https://doi.org/10.1155/2009/421425
  12. 12.
    Kingma DP, Welling M (2014) Auto-encoding variational bayes. In: 2nd international conference on learning and representant ICLR 2014—conference track process, no. Ml, pp 1–14Google Scholar
  13. 13.
    Mescheder L, Nowozin S, Geiger A (2017) Adversarial variational bayes: unifying variational autoencoders and generative adversarial networks. In: 34th international conference on machine and learning. ICML 2017, vol 5, pp 3694–3707Google Scholar
  14. 14.
    Li X, She J (2017) Collaborative variational autoencoder for recommender systems. In: Proceedings of ACM SIGKDD international conference on knowledge discovery data mining, vol. Part F1296, pp 305–314.  https://doi.org/10.1145/3097983.3098077
  15. 15.

Copyright information

© Springer Nature Singapore Pte Ltd. 2022

Authors and Affiliations

  1. 1.Laboratory of Intelligent Systems, Georesources and Renewable Energies (SIGER)University Sidi Mohamed Ben AbdellahFezMorocco

Personalised recommendations