{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T21:17:14Z","timestamp":1772054234868,"version":"3.50.1"},"reference-count":32,"publisher":"China Science Publishing & Media Ltd.","issue":"3","license":[{"start":{"date-parts":[[2023,2,22]],"date-time":"2023-02-22T00:00:00Z","timestamp":1677024000000},"content-version":"vor","delay-in-days":52,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2023,8,1]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n               <jats:p>Recently, convolutional neural networks (CNNs) have achieved excellent performance for the recommendation system by extracting deep features and building collaborative filtering models. However, CNNs have been verified susceptible to adversarial examples. This is because adversarial samples are subtle non-random disturbances, which indicates that machine learning models produce incorrect outputs. Therefore, we propose a novel model of Adversarial Neural Collaborative Filtering with Embedding Dimension Correlations, named ANCF in short, to address the adversarial problem of CNN-based recommendation system. In particular, the proposed ANCF model adopts the matrix factorization to train the adversarial personalized ranking in the prediction layer. This is because matrix factorization supposes that the linear interaction of the latent factors, which are captured between the user and the item, can describe the observable feedback, thus the proposed ANCF model can learn more complicated representation of their latent factors to improve the performance of recommendation. In addition, the ANCF model utilizes the outer product instead of the inner product or concatenation to learn explicitly pairwise embedding dimensional correlations and obtain the interaction map from which CNNs can utilize its strengths to learn high-order correlations. As a result, the proposed ANCF model can improve the robustness performance by the adversarial personalized ranking, and obtain more information by encoding correlations between different embedding layers. Experimental results carried out on three public datasets demonstrate that the ANCF model outperforms other existing recommendation models.<\/jats:p>","DOI":"10.1162\/dint_a_00151","type":"journal-article","created":{"date-parts":[[2023,2,22]],"date-time":"2023-02-22T12:35:11Z","timestamp":1677069311000},"page":"786-806","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":3,"title":["Adversarial Neural Collaborative Filtering with Embedding Dimension Correlations"],"prefix":"10.3724","volume":"5","author":[{"given":"Yi","family":"Gao","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianxia","family":"Chen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Liang","family":"Xiao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hongyang","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Liwei","family":"Pan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuan","family":"Wen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhiwei","family":"Ye","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xinyun","family":"Wu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"2026","published-online":{"date-parts":[[2023,8,1]]},"reference":[{"key":"2023091215375876500_ref1","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1145\/3159652.3159727","article-title":"Latent cross: Making use of context in recurrent Recommendation systems","volume-title":"Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining","author":"Beutel","year":"2018"},{"key":"2023091215375876500_ref2","first-page":"2227","article-title":"Outer product-based neural collaborative filtering","volume-title":"Proceedings of the 27th International Joint Conference on Artificial Intelligence","author":"He","year":"2018"},{"key":"2023091215375876500_ref3","article-title":"BPR: Bayesian personalized ranking from implicit feedback","volume-title":"arXiv preprint arXiv:1205.2618","author":"Rendle","year":"2012"},{"key":"2023091215375876500_ref4","first-page":"726","article-title":"Convolutional neural collaborative filtering with stacked embeddings","volume-title":"Asian Conference on Machine Learning","author":"Han","year":"2019"},{"key":"2023091215375876500_ref5","article-title":"Explaining and harnessing adversarial examples","volume-title":"arXiv preprint arXiv:1412.6572","author":"Goodfellow","year":"2014"},{"key":"2023091215375876500_ref6","first-page":"1765","article-title":"Universal adversarial perturbations","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Moosavi-Dezfooli","year":"2017"},{"key":"2023091215375876500_ref7","first-page":"355","article-title":"Adversarial personalized ranking for recommendation","volume-title":"The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval","author":"He","year":"2018"},{"key":"2023091215375876500_ref8","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1145\/3038912.3052569","article-title":"Neural collaborative filtering","volume-title":"Proceedings of the 26th International Conference on World Wide Web","author":"He","year":"2017"},{"issue":"1","key":"2023091215375876500_ref9","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v30i1.9973","article-title":"VBPR: visual bayesian personalized ranking from implicit feedback","volume":"Vol. 30","author":"He","year":"2016","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"2023091215375876500_ref10","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1145\/2835776.2835837","article-title":"Collaborative denoising auto-encoders for top-n Recommendation systems","volume-title":"Proceedings of the Ninth ACM International Conference on Web Search and Data Mining","author":"Wu","year":"2016"},{"key":"2023091215375876500_ref11","first-page":"3203","article-title":"Deep Matrix Factorization Models for Recommendation systems","volume":"17","author":"Xue","year":"2017","journal-title":"IJCAI"},{"key":"2023091215375876500_ref12","first-page":"335","article-title":"Attentive collaborative filtering: Multimedia recommendation with item-and component-level attention","volume-title":"Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Chen","year":"2017"},{"key":"2023091215375876500_ref13","doi-asserted-by":"crossref","first-page":"582","DOI":"10.1145\/3289600.3290975","article-title":"A simple convolutional generative network for next item recommendation","volume-title":"Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining","author":"Yuan","year":"2019"},{"key":"2023091215375876500_ref14","doi-asserted-by":"crossref","first-page":"1449","DOI":"10.1145\/3132847.3132892","article-title":"Joint representation learning for top-n recommendation with heterogeneous information sources","volume-title":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","author":"Zhang","year":"2017"},{"issue":"4","key":"2023091215375876500_ref15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3357154","article-title":"Modeling embedding dimension correlations via convolutional neural collaborative filtering","volume":"37","author":"Du","year":"2019","journal-title":"ACM Transactions on Information Systems (TOIS)"},{"key":"2023091215375876500_ref16","article-title":"Multi-Step Adversarial Perturbations on Recommendation systems Embeddings","volume-title":"arXiv preprint arXiv:2010.01329","author":"Anelli","year":"2020"},{"key":"2023091215375876500_ref17","article-title":"Adversarial machine learning at scale","volume-title":"arXiv preprint arXiv:1611. 01236","author":"Kurakin","year":"2016"},{"issue":"5","key":"2023091215375876500_ref18","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1109\/TKDE.2019.2893638","article-title":"Adversarial training towards robust multimedia recommender system","volume":"32","author":"Tang","year":"2019","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"2023091215375876500_ref19","doi-asserted-by":"crossref","DOI":"10.24963\/ijcai.2019\/187","article-title":"Deep adversarial social recommendation","volume-title":"arXiv preprint arXiv:1905.13160","author":"Fan","year":"2019"},{"key":"2023091215375876500_ref20","article-title":"Adversarial examples in modern machine learning: A review","volume-title":"arXiv preprint arXiv:1911.05268","author":"Wiyatno","year":"2019"},{"key":"2023091215375876500_ref21","doi-asserted-by":"crossref","first-page":"2467","DOI":"10.1145\/3219819.3220004","article-title":"Neural memory streaming recommender networks with adversarial training","volume-title":"Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","author":"Wang","year":"2018"},{"key":"2023091215375876500_ref22","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1145\/3240323.3240383","article-title":"RecGAN: recurrent generative adversarial networks for recommendation systems","volume-title":"Proceedings of the 12th ACM Conference on Recommendation Systems","author":"Bharadhwaj","year":"2018"},{"key":"2023091215375876500_ref23","first-page":"515","article-title":"Irgan: A minimax game for unifying generative and discriminative information retrieval models","volume-title":"Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Wang","year":"2017"},{"key":"2023091215375876500_ref24","doi-asserted-by":"crossref","first-page":"2616","DOI":"10.1145\/3308558.3313413","article-title":"Rating augmentation with generative adversarial networks towards accurate collaborative filtering","volume-title":"The World Wide Web Conference","author":"Chae","year":"2019"},{"key":"2023091215375876500_ref25","first-page":"3676","article-title":"PLASTIC: Prioritize Long and Short-term Information in Top-n Recommendation using Adversarial Training","volume-title":"Ijcai","author":"Zhao","year":"2018"},{"issue":"3","key":"2023091215375876500_ref26","doi-asserted-by":"crossref","first-page":"555","DOI":"10.1109\/TMM.2018.2887018","article-title":"Enhancing the robustness of neural collaborative filtering systems under malicious attacks","volume":"21","author":"Du","year":"2018","journal-title":"IEEE Transactions on Multimedia"},{"key":"2023091215375876500_ref27","article-title":"Neural network matrix factorization","volume-title":"arXiv preprint arXiv:1511.06443","author":"Dziugaite","year":"2015"},{"key":"2023091215375876500_ref28","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1145\/3269206.3271743","article-title":"Cfgan: A generic collaborative filtering framework based on generative adversarial networks","volume-title":"Proceedings of the 27th ACM International Conference on Information and Knowledge Management","author":"Chae","year":"2018"},{"key":"2023091215375876500_ref29","doi-asserted-by":"crossref","first-page":"659","DOI":"10.1145\/2487575.2487589","article-title":"Fism: factored item similarity models for top-n Recommendation systems","volume-title":"Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Kabbur","year":"2013"},{"key":"2023091215375876500_ref30","doi-asserted-by":"crossref","first-page":"426","DOI":"10.1145\/1401890.1401944","article-title":"Factorization meets the neighborhood: a multifaceted collaborative filtering model","volume-title":"Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Koren","year":"2008"},{"key":"2023091215375876500_ref31","first-page":"77","article-title":"The TREC-8 question answering track report","volume":"99","author":"Voorhees","year":"1999","journal-title":"Trec"},{"issue":"4","key":"2023091215375876500_ref32","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1145\/582415.582418","article-title":"Cumulated gain-based evaluation of IR techniques","volume":"20","author":"J\u00e4rvelin","year":"2002","journal-title":"ACM Transactions on Information Systems (TOIS)"}],"container-title":["Data Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/dint\/article-pdf\/5\/3\/786\/2158266\/dint_a_00151.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/dint\/article-pdf\/5\/3\/786\/2158266\/dint_a_00151.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,4]],"date-time":"2025-03-04T23:06:22Z","timestamp":1741129582000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/dint\/article\/5\/3\/786\/114954\/Adversarial-Neural-Collaborative-Filtering-with"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023]]},"references-count":32,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2023,8,1]]}},"URL":"https:\/\/doi.org\/10.1162\/dint_a_00151","relation":{},"ISSN":["2641-435X"],"issn-type":[{"value":"2641-435X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023]]},"published":{"date-parts":[[2023]]}}}