Sequential Recommendation with Self-Attentive Multi-Adversarial Network

SIGIR '20: The 43rd International ACM SIGIR conference on research and development in Information Retrieval Virtual Event China July, 2020, pp. 89-98, 2020.

Cited by: 0|Bibtex|Views148|DOI:https://doi.org/10.1145/3397271.3401111
EI
Other Links: arxiv.org|dl.acm.org|dblp.uni-trier.de|academic.microsoft.com
Weibo:
We have proposed a Multi-Factor Generative Adversarial Network for sequential recommendation

Abstract:

Recently, deep learning has made significant progress in the task of sequential recommendation. Existing neural sequential recommenders typically adopt a generative way trained with Maximum Likelihood Estimation (MLE). When context information (called factor) is involved, it is difficult to analyze when and how each individual factor woul...More

Code:

Data:

0
Introduction
  • Recommender systems aim to accurately characterize user interests and provide personalized recommendations in a variety of real-world applications.
  • They serve as an important information filtering technique to alleviate the information overload problem and enhance user experiences.
  • Several studies have proposed to incorporate context information to enhance the performance of neural sequential recommenders [11, 12, 16]
  • The advantages of these sequential neural networks have been experimentally confirmed as they have achieved significant performance improvements
Highlights
  • Recommender systems aim to accurately characterize user interests and provide personalized recommendations in a variety of real-world applications
  • In this paper, we present a novel Multi-Factor Generative Adversarial Network (MFGAN)
  • GRUF further improves over the Gated Recurrent Units (GRU) method, indicating context information is useful in our task
  • Self-attention architecture has shown its superiority in various tasks [5, 27]. Such an architecture is useful when dealing with sequence data, which works well in sequential recommendation
  • We have proposed a Multi-Factor Generative Adversarial Network (MFGAN) for sequential recommendation
  • We propose several techniques to improve the performance of our approach
  • We have found that using multiple discriminators is useful to stabilize the training of adversarial learning, and enhance the interpretability of recommendation algorithms
Methods
  • The authors compare the propose approach MFGAN against a number of competitive baselines.
Results
  • Results and Analysis

    The results of different methods for sequential recommendation are presented in Table 2.
  • (2) Among sequential recommendation baselines, the Markov Chain-based method performs better on sparse dataset than dense datasets.
  • Self-attention architecture has shown its superiority in various tasks [5, 27].
  • Such an architecture is useful when dealing with sequence data, which works well in sequential recommendation
Conclusion
  • Discussion and Analysis

    the authors analyze the effectiveness and the stability of the multi-adversarial architecture in the task.

    As mentioned before, the authors train the MFGAN model in an RL way by policy gradient.
  • Since the authors have multiple discriminators, from each discriminator the authors receive a reward to guide the training process of the generator.
  • Recall that the authors use a λ-parameterized softmax function to combine the reward signals from multiple discriminators in Eq (10).
  • The low reward only indicates the position where to decrease pG (x), and does not indicate the position where to increase pG (x).In this paper, the authors have proposed a Multi-Factor Generative Adversarial Network (MFGAN) for sequential recommendation.
  • The authors will consider incorporating explicit user preference in the discriminators
Summary
  • Introduction:

    Recommender systems aim to accurately characterize user interests and provide personalized recommendations in a variety of real-world applications.
  • They serve as an important information filtering technique to alleviate the information overload problem and enhance user experiences.
  • Several studies have proposed to incorporate context information to enhance the performance of neural sequential recommenders [11, 12, 16]
  • The advantages of these sequential neural networks have been experimentally confirmed as they have achieved significant performance improvements
  • Methods:

    The authors compare the propose approach MFGAN against a number of competitive baselines.
  • Results:

    Results and Analysis

    The results of different methods for sequential recommendation are presented in Table 2.
  • (2) Among sequential recommendation baselines, the Markov Chain-based method performs better on sparse dataset than dense datasets.
  • Self-attention architecture has shown its superiority in various tasks [5, 27].
  • Such an architecture is useful when dealing with sequence data, which works well in sequential recommendation
  • Conclusion:

    Discussion and Analysis

    the authors analyze the effectiveness and the stability of the multi-adversarial architecture in the task.

    As mentioned before, the authors train the MFGAN model in an RL way by policy gradient.
  • Since the authors have multiple discriminators, from each discriminator the authors receive a reward to guide the training process of the generator.
  • Recall that the authors use a λ-parameterized softmax function to combine the reward signals from multiple discriminators in Eq (10).
  • The low reward only indicates the position where to decrease pG (x), and does not indicate the position where to increase pG (x).In this paper, the authors have proposed a Multi-Factor Generative Adversarial Network (MFGAN) for sequential recommendation.
  • The authors will consider incorporating explicit user preference in the discriminators
Tables
  • Table1: Statistics of the three datasets. Dataset #Users #Items #Interactions #Factors
  • Table2: Performance comparison of different methods for sequential recommendation task on three datasets. We use bold and underline fonts to denote the best performance and second best performance method in each metric respectively
  • Table3: Variant comparisons of our MFGAN framework on Movielens-1m dataset
Download tables as Excel
Related work
  • In this section, we review studies closely related to our work in two aspects.

    Sequential Recommendation. Early works on sequential recommendation are mainly based on Markov Chain (MC) assumption. For instance, Rendle et al [26] fuse the matrix factorization and firstorder Markov Chain for modeling global user preference and shortterm interests, respectively. Another line to model user behaviors is resorting to the recurrent neural network, which has achieved great success on sequential modeling in a variety of applications. Hidasi et al [10] firstly introduce Gated Recurrent Units (GRU) to the sessionbased recommendation. A surge of following variants modify it by introducing pair-wise loss function [11], attention mechanism [18], memory network [3], hierarchical structure [22], etc. Other architectures or networks have also been used [19, 20], achieving good performance. Moreover, context information is often used to improve the recommendation performance and interpretability [11,12,13, 29]. Recently, self-attention network has achieved significant improvement in a bunch of NLP tasks [5, 27] and inspired a new direction on applying the self-attention mechanism to sequential recommendation problem [14].
Funding
  • This work was partially supported by the National Natural Science Foundation of China under Grant No 61872369 and 61832017, the Fundamental Research Funds for the Central Universities, the Research Funds of Renmin University of China under Grant No.18XNL G22 and 19XNQ047, Beijing Academy of Artificial Intelligence (BAAI) under Grant No BAAI2020ZJ0301, and Beijing Outstanding Young Scientist Program under Grant No BJJWZYJH012019100020098. Xin Zhao is the corresponding author
Reference
  • Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Oksana Yakhnenko. 2013. Translating embeddings for modeling multi-relational data. In Advances in neural information processing systems. 2787–2795.
    Google ScholarFindings
  • Dong-Kyu Chae, Jin-Soo Kang, Sang-Wook Kim, and Jung-Tae Lee. 2018. Cfgan: A generic collaborative filtering framework based on generative adversarial networks. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management. 137–146.
    Google ScholarLocate open access versionFindings
  • Xu Chen, Hongteng Xu, Yongfeng Zhang, Jiaxi Tang, Yixin Cao, Zheng Qin, and Hongyuan Zha. 2018. Sequential recommendation with user memory networks. In Proceedings of the eleventh ACM International Conference on Web Search and Data Mining. 108–116.
    Google ScholarLocate open access versionFindings
  • Kyunghyun Cho, Bart van Merrienboer, Çaglar Gülçehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 201Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. 1724–1734.
    Google ScholarLocate open access versionFindings
  • Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 4171–4186.
    Google ScholarLocate open access versionFindings
  • Ishan P. Durugkar, Ian Gemp, and Sridhar Mahadevan. 2017. Generative MultiAdversarial Networks. In 5th International Conference on Learning Representations.
    Google ScholarLocate open access versionFindings
  • Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In Advances in neural information processing systems. 2672–2680.
    Google ScholarFindings
  • Google. 2016. Freebase Data Dumps. https://developers.google.com/freebase/data.
    Findings
  • F Maxwell Harper and Joseph A Konstan. 2015. The movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis) 5, 4 (2015), 1–19.
    Google ScholarFindings
  • Balázs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, and Domonkos Tikk. 2016. Session-based Recommendations with Recurrent Neural Networks. In 4th International Conference on Learning Representations.
    Google ScholarLocate open access versionFindings
  • Balázs Hidasi, Massimo Quadrana, Alexandros Karatzoglou, and Domonkos Tikk. 2016. Parallel recurrent neural network architectures for feature-rich session-based recommendations. In Proceedings of the 10th ACM Conference on Recommender Systems. 241–248.
    Google ScholarLocate open access versionFindings
  • Jin Huang, Zhaochun Ren, Wayne Xin Zhao, Gaole He, Ji-Rong Wen, and Daxiang Dong. 2019. Taxonomy-aware multi-hop reasoning networks for sequential recommendation. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining. 573–581.
    Google ScholarLocate open access versionFindings
  • Jin Huang, Wayne Xin Zhao, Hongjian Dou, Ji-Rong Wen, and Edward Y Chang.
    Google ScholarFindings
  • 2018. Improving sequential recommendation with knowledge-enhanced memory networks. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. 505–514.
    Google ScholarLocate open access versionFindings
  • [14] Wang-Cheng Kang and Julian McAuley. 2018. Self-attentive sequential recommendation. In 2018 IEEE International Conference on Data Mining (ICDM). IEEE, 197–206.
    Google ScholarLocate open access versionFindings
  • [15] Christian Ledig, Lucas Theis, Ferenc Huszar, Jose Caballero, Andrew Cunningham, Alejandro Acosta, Andrew P. Aitken, Alykhan Tejani, Johannes Totz, Zehan Wang, and Wenzhe Shi. 2017. Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. In 2017 IEEE Conference on Computer Vision and Pattern Recognition. 105–114.
    Google ScholarLocate open access versionFindings
  • [16] Chenliang Li, Xichuan Niu, Xiangyang Luo, Zhenzhong Chen, and Cong Quan. 2019. A Review-Driven Neural Model for Sequential Recommendation. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. 2866–2872.
    Google ScholarLocate open access versionFindings
  • [17] Chun-Liang Li, Wei-Cheng Chang, Yu Cheng, Yiming Yang, and Barnabás Póczos.
    Google ScholarFindings
  • 2017. Mmd gan: Towards deeper understanding of moment matching network. In Advances in neural information processing systems. 2203–2213.
    Google ScholarFindings
  • [18] Jing Li, Pengjie Ren, Zhumin Chen, Zhaochun Ren, Tao Lian, and Jun Ma. 2017. Neural Attentive Session-based Recommendation. In Proceedings of the 2017 ACM on Conference on Information and Knowledge Management. 1419–1428.
    Google ScholarLocate open access versionFindings
  • [19] Fuyu Lv, Taiwei Jin, Changlong Yu, Fei Sun, Quan Lin, Keping Yang, and Wilfred Ng. 2019. SDM: Sequential deep matching model for online large-scale recommender system. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 2635–2643.
    Google ScholarLocate open access versionFindings
  • [20] Chen Ma, Peng Kang, and Xue Liu. 2019. Hierarchical gating networks for sequential recommendation. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 825–833.
    Google ScholarLocate open access versionFindings
  • [21] Sebastian Nowozin, Botond Cseke, and Ryota Tomioka. 2016. f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization. In Advances in neural information processing systems. 271–279.
    Google ScholarLocate open access versionFindings
  • [22] Massimo Quadrana, Alexandros Karatzoglou, Balázs Hidasi, and Paolo Cremonesi.
    Google ScholarFindings
  • 2017. Personalizing Session-based Recommendations with Hierarchical Recurrent Neural Networks. In Proceedings of the Eleventh ACM Conference on Recommender Systems. 130–137.
    Google ScholarLocate open access versionFindings
  • [23] Marc’Aurelio Ranzato, Sumit Chopra, Michael Auli, and Wojciech Zaremba. 2016. Sequence Level Training with Recurrent Neural Networks. 4th International Conference on Learning Representations (2016).
    Google ScholarFindings
  • [24] Steffen Rendle. 2012. Factorization machines with libfm. ACM Transactions on Intelligent Systems and Technology (TIST) 3, 3 (2012), 1–22.
    Google ScholarLocate open access versionFindings
  • [25] Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme. 2009. BPR: Bayesian Personalized Ranking from Implicit Feedback. In UAI.
    Google ScholarLocate open access versionFindings
  • [26] Steffen Rendle, Christoph Freudenthaler, and Lars Schmidt-Thieme. 2010. Factorizing personalized markov chains for next-basket recommendation. In Proceedings of the 19th International World Wide Web Conference. 811–820.
    Google ScholarLocate open access versionFindings
  • [27] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. In Advances in neural information processing systems. 5998–6008.
    Google ScholarFindings
  • [28] Jun Wang, Lantao Yu, Weinan Zhang, Yu Gong, Yinghui Xu, Benyou Wang, Peng Zhang, and Dell Zhang. 2017. Irgan: A minimax game for unifying generative and discriminative information retrieval models. In Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. 515–524.
    Google ScholarLocate open access versionFindings
  • [29] Shaoqing Wang, Cuiping Li, Kankan Zhao, and Hong Chen. 2017. Context-aware recommendations with random partition factorization machines. Data Science and Engineering 2, 2 (2017), 125–135.
    Google ScholarLocate open access versionFindings
  • [30] Jiqing Wu, Zhiwu Huang, Janine Thoma, Dinesh Acharya, and Luc Van Gool. 2018. Wasserstein divergence for gans. In Proceedings of the European Conference on Computer Vision (ECCV). 653–668.
    Google ScholarLocate open access versionFindings
  • [31] Qiong Wu, Yong Liu, Chunyan Miao, Binqiang Zhao, Yin Zhao, and Lu Guan.
    Google ScholarFindings
  • 2019. PD-GAN: Adversarial Learning for Personalized Diversity-Promoting Recommendation. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. 3870–3876.
    Google ScholarLocate open access versionFindings
  • [32] Lantao Yu, Weinan Zhang, Jun Wang, and Yong Yu. 2017. Seqgan: Sequence generative adversarial nets with policy gradient. In Thirty-First AAAI Conference on Artificial Intelligence.
    Google ScholarLocate open access versionFindings
  • [33] Yizhe Zhang, Zhe Gan, Kai Fan, Zhi Chen, Ricardo Henao, Dinghan Shen, and Lawrence Carin. 2017. Adversarial Feature Matching for Text Generation. In Proceedings of the 34th International Conference on Machine Learning. 4006–4015.
    Google ScholarLocate open access versionFindings
  • [34] Wayne Xin Zhao, Gaole He, Kunlin Yang, Hong-Jian Dou, Jin Huang, Siqi Ouyang, and Ji-Rong Wen. 2019. KB4Rec: A Data Set for Linking Knowledge Bases with Recommender Systems. Data Intelligence 1, 2 (2019), 121–136.
    Google ScholarLocate open access versionFindings
Full Text
Your rating :
0

 

Tags
Comments