基于自监督学习和数据回放的新闻推荐模型增量学习方法

doi:10.3969/j.issn.1006-2475.2023.12.001

摘要/Abstract

摘要： 摘要：个性化的新闻推荐技术对于缓解信息过载和提高用户体验具有重要意义。新闻推荐模型通常基于固定数据集进行迭代训练，然而在现实场景下，新闻推荐模型需要不断学习以适应新的用户和新闻。为此，增量学习被提出用于帮助新闻推荐模型进行增量更新。新闻推荐模型增量学习的主要挑战为灾难性遗忘问题，即模型会忘记它之前学习过的用户偏好。鉴于此，本文提出基于自监督学习和数据回放（Self-supervised Learning and Data Replay）的新闻推荐模型增量学习方法SSL-DR。SSL-DR首先在新闻推荐任务中加入自监督学习任务来获取用户的稳定偏好，有效减轻灾难性遗忘问题。为了巩固已学知识，SSL-DR进一步通过基于用户对于候选新闻的点击概率分数的采样策略实现数据回放，同时借助知识蒸馏策略转移已学知识。实验结果表明，本文方法可有效提升新闻推荐模型在增量训练过程中的整体推荐性能，明显缓解灾难性遗忘问题。

Abstract: Abstract： Personalized news recommendation technology is important to alleviate information overload and improve user experience. News recommendation models are usually iteratively trained based on fixed data sets. However， in real scenarios， news recommendation models need to constantly learn to adapt to new users and news. Therefore， incremental learning has been proposed to help models perform incremental updates. The main challenge of the incremental learning of news recommendation models is the catastrophic forgetting problem， where a trained model forgets the user preferences it has previously learned. In view of this， this paper proposes SSL-DR， an incremental learning method of news recommendation models based on self-supervised learning and data replay. SSL-DR firstly adds the self-supervised learning task to the news recommendation task to obtain the user's stable preference， which effectively reduces the problem of catastrophic forgetting. To consolidate the learned knowledge， SSL-DR further implements a sampling strategy based on the user's click probability scores for candidate news to achieve data replay and transfer the learned knowledge through a knowledge distillation strategy. The experimental results show that， our method can effectively improve the overall recommendation performance of the news recommendation model in the process of incremental training， and significantly alleviate the problem of catastrophic forgetting.

Key words: Key words： news recommendation, incremental learning, catastrophic forgetting, self-supervised learning, deep learning

中图分类号:

TP311

林威. 基于自监督学习和数据回放的新闻推荐模型增量学习方法[J]. 计算机与现代化, 2023, 0(12): 1-6.

LIN Wei. Incremental News Recommendation Method Based on Self-supervised Learning and Data Replay[J]. Computer and Modernization, 2023, 0(12): 1-6.

参考文献［26］

［1］	LI M M， WANG L C. A survey on personalized news recommendation technology［J］. IEEE Access， 2019，7：145861
	-145879.
［2］	FRENCH R M. Catastrophic forgetting in connectionist networks［J］. Trends in Cognitive Sciences， 1999，3（4）：128-135.
［3］	DEVLIN J， CHANG M W， LEE K， et al. BERT： Pre-training of deep bidirectional transformers for language understanding［C］// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics： Human Language Technologies. 2019：4171-4186.
［4］	WU C H， WU F Z， QI T， et al. Empowering news recommendation with pre-trained language models［C］// Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021：1652-1656.
［5］	CHEN T， KORNBLITH S， NOROUZI M， et al. A simple framework for contrastive learning of visual representations［C］// Proceedings of the 37th International Conference on Machine Learning. 2020：1597-1607.
［6］	HINTON G， VINYALS O， DEAN J. Distilling the knowledge in a neural network［J］. arXiv preprint arXiv：1503.02531， 2015.
［7］	WU F Z， QIAO Y， CHEN J H， et al. MIND： A large-scale dataset for news recommendation［C］// Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020：3597-3606.
［8］	SON J W， KIM A Y， PARK S B. A location-based news article recommendation with explicit localized semantic analysis［C］// Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2013：293-302.
［9］	ZHU Q N， ZHOU X F， SONG Z L， et al. DAN： Deep attention neural network for news recommendation［C］// Proceedings of the 2019 AAAI Conference on Artificial Intelligence. 2019，33（1）：5973-5980.
［10］	WU C H， WU F Z， AN M X， et al. NPA： Neural news recommendation with personalized attention［C］// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019：2576-2584.
［11］	ZHANG H J， SHEN Z D. News recommendation based on user topic and entity preferences in historical behavior［J］. Information， 2023，14（2）. DOI： 10.3390/info14020060.
［12］	VAN De VEN G M， TOLIAS A S. Three scenarios for continual learning［J］. arXiv preprint arXiv：1904.07734， 2019.
［13］	LI Z Z， HOIEM D. Learning without forgetting［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2018，40（12）：2935-2947.
［14］	KIRKPATRICK J， PASCANU R， RABINOWITZ N， et al. Overcoming catastrophic forgetting in neural networks［J］. Proceedings of the National Academy of Sciences of the United States of America， 2017，114（13）：3521-3526.
［15］	ZENKE F， POOLE B， GANGULI S. Continual learning through synaptic intelligence［C］// Proceedings of the 34th International Conference on Machine Learning. 2017：3987-3995.
［16］	CHAUDHRY A， DOKANIA P K， AJANTHAN T， et al. Riemannian walk for incremental learning： Understanding forgetting and intransigence［C］// Proceedings of the 2018 European Conference on Computer Vision （ECCV）. 2018：532-547.
［17］	REBUFFI S A， KOLESNIKOV A， SPERL G， et al. iCaRL： Incremental classifier and representation learning［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. 2017：5533-5542.
［18］	RIEMER M， CASES I， AJEMIAN R， et al. Learning to learn without forgetting by maximizing transfer and minimizing interference［J］. arXiv preprint arXiv：1810.11910， 2018.
［19］	BUZZEGA P， BOSCHINI M， PORRELLO A， et al. Dark experience for general continual learning： A strong， simple baseline［C］// Proceedings of the 2020 International Conference on Neural Information Processing Systems. 2020：15920-15930.
［20］	ERICSSON L， GOUK H， LOY C C， et al. Self-supervised representation learning： Introduction， advances， and challenges［J］. IEEE Signal Processing Magazine， 2022，39（3）：42-62.
［21］	HE K M， FAN H Q， WU Y X， et al. Momentum contrast for unsupervised visual representation learning［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. 2020：9726-9735.
［22］	GRILL J B， STRUB F， ALTCHE F， et al. Bootstrap your own latent： A new approach to self-supervised learning［C］// Proceedings of the 2020 International Conference on Neural Information Processing Systems. 2020：21271-21284.
［23］	CHEN X L， HE K M. Exploring simple Siamese representation learning［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. 2021：15745-15753.
［24］	WU C H， WU F Z， QI T， et al. Is news recommendation a sequential recommendation task［C］// Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2022：2382-2386.
［25］	BARDES A， PONCE J， LECUN Y. VICReg： Variance-invariance-covariance regularization for self-supervised learning［J］. arXiv preprint arXiv：2105.04906， 2021.
［26］	ZBONTAR J， JING L， MISRA I， et al. Barlow twins： Self-supervised learning via redundancy reduction［C］// Proceedings of the 38th International Conference on Machine Learning. 2021：12310-12320.

[1]	乔善宝, 高永彬, 黄勃, 余文俊. 基于条件卷积与极化自注意力的单目深度与位姿估计[J]. 计算机与现代化, 2023, 0(02): 34-39.
[2]	王梦, 张鸿鑫, 刘庆华, 张东. 基于改进YOLOv5的幽门螺杆菌免疫印迹图像识别[J]. 计算机与现代化, 2022, 0(09): 78-84.
[3]	李芬1，肖建2，林志强2，李志鹏1 . 伴随时空特性的雷电预测BP-ANN模型研究[J]. 计算机与现代化, 2019, 0(04): 76-.
[4]	崔丽娜1，郭虎升2. 基于增量学习的非平衡SVM分类方法[J]. 计算机与现代化, 2018, 0(07): 20-.
[5]	孙震，王敬东，茅天诒，魏雪迎. 基于自监督学习的河流分割方法[J]. 计算机与现代化, 2017, 0(10): 10-14.
[6]	李金华;梁永全;吕芳芳. 一种加权朴素贝叶斯分类增量学习模型[J]. 计算机与现代化, 2010, 1(5): 30-32.
[7]	刘建军;胡卫东;郁文贤. 基于前馈神经网络的增量学习研究[J]. 计算机与现代化, 2009, 1(7): 1-4,43.