融合多维梯度反馈的生成对抗网络推荐系统

doi:10.3778/j.issn.1673-9418.2303039

摘要/Abstract

摘要： 互联网时代，推荐系统在日常生活中变得十分重要，生成对抗网络（GAN）与推荐算法的结合为该领域的发展提供了新契机。以往基于生成对抗网络的推荐系统中，鉴别器提供的梯度反馈是二元的，此类反馈为生成器提供的帮助不够全面，造成诸如生成器性能不稳定、迭代速度慢等问题，进而影响模型的整体推荐效果。针对此问题，提出了多维梯度反馈生成对抗网络（MGFGAN），根据生成器生成的多维用户评分向量，该模型将自编码器（AutoEncoder）融入鉴别器中，达到为生成器提供多元反馈的目的，旨在让生成器生成的数据更加贴近用户偏好；此外，融合多维梯度反馈机制给模型整体带来了运算量激增的问题；因此，MGFGAN在生成器中引入了负采样模块，使得生成器同时兼顾用户感兴趣和不感兴趣的物品，从而加速生成器快速生成与真实用户分布一致的数据，提升模型的效率。提出的模型在公开数据集FilmTrust和Ciaos上展开实验仿真。实验结果表明MGFGAN的推荐性能优于其他基于生成对抗网络的推荐模型，并且在效率和稳定性方面取得改善。

Abstract: In the Internet era, recommender systems become more and more significant in the daily life. The combination of generative adversarial networks (GAN) and recommended algorithm provides new opportunities for the development of this field. In previous recommendation systems based on GAN, the gradient feedback provided by the discriminator is binary, which does not comprehensively assist the generator. This inadequacy leads to issues such as unstable generator performance and slow model iteration speed, thereby reducing the overall effectiveness of recommendations. Multi-dimensional gradient feedback generative adversarial networks (MGFGAN) is proposed to address above problems. According to the type of generated multidimensional user rating vector, the model incorporates AutoEncoder in the discriminator to provide more diversified feedback for the generator, aiming to make the generated data more closely match the user’s preferences of the model. However, it brings the problem of increasing computational complexity to the model. Therefore, MGFGAN introduces a negative sampling module in the generator, which makes the generator take into account both items of interest and disinterest to the user, thus accelerating the generator to quickly generate data consistent with the real user distribution and improving the efficiency of the model. Finally, the MGFGAN is carried out experimental simulation on the public datasets FilmTrust and Ciaos. Experimental results show that the recommendation performance of MGFGAN outperforms other recommendation models based on GAN and achieves improvements in efficiency and stability.

Key words: recommendation system, multi-dimensional gradient feedback, generative adversarial networks (GAN), collaborative filtering

李祥霞, 陈楷锐, 李彬. 融合多维梯度反馈的生成对抗网络推荐系统[J]. 计算机科学与探索, 2024, 18(6): 1579-1589.

LI Xiangxia, CHEN Kairui, LI Bin. Generative Adversarial Network Recommendation System with Multi-dimensional Gradient Feedback Mechanism[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(6): 1579-1589.

参考文献

[1] LEE J, LEE D, LEE Y C, et al. Improving the accuracy of top-N recommendation using a preference model[J]. Information Sciences, 2016, 348: 290-304.
[2] DAVIDSON J, LIEBALD B, LIU J, et al. The YouTube video recommendation system[C]//Proceedings of the 4th ACM Conference on Recommender Systems, Barcelona, Sep 26-30, 2010. New York: ACM, 2010: 293-296.
[3] SALAKHUTDINOV R, MNIH A, HINTON G. Restricted Boltzmann machines for collaborative filtering[C]//Proceedings of the 24th International Conference on Machine Learning, Corvallis, Jun 20-24, 2007. New York: ACM, 2007: 791-798.
[4] JOHNSON C C. Logistic matrix factorization for implicit feedback data[C]//Advances in Neural Information Processing Systems 27, Montreal, Dec 8-13, 2014: 1-9.
[5] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. arXiv:1406.2661, 2014.
[6] NI S, ZHOU W, WEN J, et al. Enhancing sequential recommendation with contrastive generative adversarial network[J]. Information Processing & Management, 2023, 60(3): 103331.
[7] 王娜, 何晓明, 刘志强, 等. 一种基于用户播放行为序列的个性化视频推荐策略[J]. 计算机学报, 2020, 43(1): 123-135.
WANG N, HE X M, LIU Z Q, et al. Personalized video recommendation strategy based on user’s playback behavior sequence[J]. Chinese Journal of Computers, 2020, 43(1): 123-135.
[8] 陈继伟, 汪海涛, 姜瑛, 等. 基于生成对抗模型的序列推荐算法[J]. 中文信息学报, 2022, 36(7): 143-153.
CHEN J W, WANG H T, JIANG Y, et al. Sequential recommendation with generative adversarial networks[J]. Journal of Chinese Information Processing, 2022, 36(7): 143-153.
[9] WANG J, YU L, ZHANG W, et al. IRGAN: a minimax game for unifying generative and discriminative information retrieval models[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, Tokyo, Aug 7-11, 2017. New York: ACM, 2017: 515-524.
[10] WANG H W, WANG J, WANG J L, et al. Graph GAN: graph representation learning with generative adversarial nets[C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence, New Orleans, Feb 2-7, 2018. Menlo Park: AAAI, 2018: 2508-2515.
[11] CHAE D K, KANG J S, KIM S W, et al. CFGAN: a generic collaborative filtering framework based on generative adversarial networks[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management, Torino, Oct 22-26, 2018. New York: ACM, 2018: 137-146.
[12] WU H, LONG J, LI N, et al. Adversarial auto-encoder domain adaptation for cold-start recommendation with positive and negative hypergraphs[J]. ACM Transactions on Information Systems, 2023, 41(2): 1-25.
[13] ZHENG Y. DGCN: diversified recommendation with graph convolutional networks[C]//Proceedings of the Web Conference 2021, Ljubljana Slovenia, Apr 19-23, 2021. New York: ACM, 2021: 401-412.
[14] CAI X, HAN J, YANG L. Generative adversarial network based heterogeneous bibliographic network representation for personalized citation recommendation[C]//Proceedings of the 2018 AAAI Conference on Artificial Intelligence, New Orleans, Feb 2-7, 2018. Menlo Park: AAAI, 2018: 5747-5754.
[15] RENDLE S, FREUDENTHALER C. Improving pairwise learning for item recommendation from implicit feedback[C]//Proceedings of the 7th ACM International Conference on Web Search and Data Mining, New York, Feb 24-28, 2014. New York: ACM, 2014: 273-282.
[16] LIANG G, ON B W, JEONG D, et al. A text GAN framework for creative essay recommendation[J]. Knowledge-Based Systems, 2021, 232: 107501.
[17] ADOMAVICIUS G, TUZHILIN A. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions[J]. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(6): 734-749.
[18] ZHANG J, FENG Y, WANG C, et al. Multi-domain clustering pruning: exploring space and frequency similarity based on GAN[J]. Neurocomputing, 2023, 542: 126279.
[19] 张玉洁, 董政, 孟祥武. 个性化广告推荐系统及其应用研究[J]. 计算机学报, 2021, 44(3): 531-563.
ZHANG Y J, DONG Z, MENG X W. Research on personalized advertising recommendation systems and their application[J]. Chinese Journal of Computers, 2021, 44(3): 531-563.
[20] GARCíA-GIL D, LUQUE-SáNCHEZ F, LUENGO J, et al. From big to smart data: iterative ensemble filter for noise filtering in big data classification[J]. International Journal of Intelligent Systems, 2019, 34(12): 3260-3274.
[21] GAO M, ZHANG J, YU J, et al. Recommender systems based on generative adversarial networks: a problem-driven perspective[J]. Information Sciences, 2021, 546: 1166-1185.
[22] SEDHAIN S, MENON A K, SANNER S, et al. AutoRec: autoencoders meet collaborative filtering[C]//Proceedings of the 24th International Conference on World Wide Web, Florence, May 18-22, 2015. New York: ACM, 2015: 111-112.
[23] WU C Y, AHMED A, BEUTEL A, et al. Recurrent recommender networks[C]//Proceedings of the 10th ACM International Conference on Web Search and Data Mining, Cambridge, Feb 6-10, 2017. New York: ACM, 2017: 495-503.
[24] WU Y, DUBOIS C, ZHENG A X, et al. Collaborative denoising auto-encoders for top-n recommender systems[C]//Proceedings of the 9th ACM International Conference on Web Search and Data Mining, San Francisco, Feb 22-25, 2016. New York: ACM, 2016: 153-162.
[25] HE X, LIAO L, ZHANG H, et al. Neural collaborative filtering[C]//Proceedings of the 26th International Conference on World Wide Web, Perth, Apr 3-7, 2017. New York: ACM, 2017: 173-182.
[26] RAZA S M, JANG B, YANG H, et al. Improved GAN with fact forcing for mobility prediction[J]. Journal of Network and Computer Applications, 2022, 207: 103488.
[27] IMRAN A S, YANG R, KASTRATI Z, et al. The impact of synthetic text generation for sentiment analysis using GAN based models[J]. Egyptian Informatics Journal, 2022, 23(3): 547-557.
[28] 吴正洋, 汤庸, 刘海. 个性化学习推荐研究综述[J]. 计算机科学与探索, 2022, 16(1): 21-40.
WU Z Y, TANG Y, LIU H. Survey of personalized learning recommendation[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(1): 21-40.
[29] BHARADHWAJ H, PARK H, LIM B Y. RecGAN: recurrent generative adversarial networks for recommendation systems[C]//Proceedings of the 12th ACM Conference on Recommender Systems, Vancouver, Oct 2-7, 2018. New York: ACM, 2018: 372-376.
[30] WANG Q, YIN H, WANG H, et al. Enhancing collaborative filtering with generative augmentation[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Anchorage, Aug 4-8, 2019. New York: ACM, 2019: 548-556.
[31] ZHAO J, MATHIEU M, LECUN Y. Energy-based generative adversarial network[J]. arXiv:1609.03126, 2016.
[32] WANG H, SHAO N, LIAN D. Adversarial binary collaborative filtering for implicit feedback[C]//Proceedings of the 2019 AAAI Conference on Artificial Intelligence, Honolulu, Jan 27-Feb 1, 2019. Menlo Park: AAAI, 2019: 5248-5255.
[33] QIN J, LIU B, QIAN J. A novel privacy-preserved recommender system framework based on federated learning[C]// Proceedings of the 4th International Conference on Software Engineering and Information Management, Beijing, Dec 10-12, 2021: 82-88.
[34] ARJOVSKY M, CHINTALA S, BOTTOU L. Wasserstein generative adversarial networks[C]//Proceedings of the 2017 International Conference on Machine Learning, Sydney, Aug 6-11, 2017: 214-223.
[35] MIRZA M, OSINDERO S. Conditional generative adversarial nets[J]. arXiv:1411.1784, 2014.
[36] 严冬梅, 李斌. 基于生成式对抗神经网络的股票预测研究[J]. 计算机工程与应用, 2022, 58(13): 185-194.
YAN D M, LI B. Research on stock prediction based on generative adversarial networks[J]. Computer Engineering and Applications, 2022, 58(13): 185-194.
[37] YANG Z, QIN J, LIN C, et al. GANRec: a negative sampling model with generative adversarial network for recommendation[J]. Expert Systems with Applications, 2023, 214: 119155.
[38] HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al. Improving neural networks by preventing co-adaptation of feature detectors[J]. arXiv:1207.0580, 2012.
[39] GUO G, ZHANG J, YORKE-SMITH N. A novel Bayesian similarity measure for recommender systems[C]//Proceedings of the 23rd International Joint Conference on Artificial Intelligence, Beijing, Aug 3-9, 2013. Palo Alto: AAAI, 2013: 2619-2625.
[40] GUO G, ZHANG J, THALMANN D, et al. Etaf: an extended trust antecedents framework for trust prediction[C]//Proceedings of the 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, Beijing, Aug 17-20, 2014. Piscataway: IEEE, 2014: 540-547.
[41] 张金柱, 蒋霖琪, 王玥, 等. 基于异构网络表示学习的相关图书推荐研究[J]. 计算机工程与应用, 2022, 58(9): 263-270.
ZHANG J Z, JIANG L Q, WANG Y, et al. Book recommendation based on heterogeneous network representation learning[J]. Computer Engineering and Applications, 2022, 58(9): 263-270.
[42] 王永贵, 林佳敏, 何佳玉. 融合领导者影响与隐式信任度的群组推荐方法[J]. 计算机工程与应用, 2022, 58(9): 98-106.
WANG Y G, LIN J M, HE J Y. Group recommendation method combining leader influence and implicit trust metrics[J]. Computer Engineering and Applications, 2022, 58(9): 98-106.
[43] YU L, ZHANG W, WANG J, et al. SeqGAN: sequence generative adversarial nets with policy gradient[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence, San Francisco, Feb 4-9, 2017. Menlo Park: AAAI, 2017: 2852-2858.
[44] DERVISHAJ E, CREMONESI P. GAN-based matrix factorization for recommender systems[C]//Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing, Gwangju, Apr 11-15, 2022. New York: ACM, 2022: 1373-1381.