recommendation collaborative distillation top-n

基于融合语义信息改进的内容推荐算法。Improved content recommendation algorithm integrating semantic information.

引言 路漫漫其修远兮,吾将上下而求索。每天一篇论文,做更好的自己。 本文读的这篇论文为发表于2023年5月28日的一篇名为《基于融合语义信息改进的内容推荐算法》(基于融合语义信息改进的内容推荐算法)的文章,文章主要介绍了基于内容的推荐技术在电子商务和教育领域的广泛应用,以及传统基于内容推荐技术在语义 ......

基于正则化的图自编码器在推荐算法中的应用 Application of graph auto-encoders based on regularization in recommendation algorithms

引言 看过的每一篇文章,都是对自己的提高。不积跬步无以至千里,不积小流无以成江海,积少成多,做更好的自己。 本文基于2023年4月6日发表于SCIPEERJ COMPUTER SCIENCE(PEERJ计算机科学)上的一篇名为《基于正则化的图自编码器在推荐算法中的应用》(Application of ......

转载:深度学习:蒸馏Distill

转载,写的比较好了,可以参考:https://blog.csdn.net/pipisorry/article/details/117257414 Distilling the knowledge in a neural networkHinton 在论文中提出方法很简单,就是让学生模型的预测分布,来 ......
深度 Distill

Top-N推荐算法 Top-N recommendation Algorithms

引言 推荐算法是计算机专业中的一种算法,通过一些计算,能够推测用户喜欢的东西,在互联网环境中应用比较广泛。Top-N算法在生活中非常常见,比如学术论文推荐论文、音乐软件推荐歌曲等。 今天看到一篇名叫"A Revisiting Study of Appropriate Offline Evaluati ......
Top-N recommendation 算法 Algorithms Top

MMGCN: Multi-modal Graph Convolution Network for Personalized Recommendation of Micro-video

目录概符号说明MMGCN代码 Wei Y., Wang X., Nie L., He X., Hong R. and Chua T. MMGCN: Multi-modal graph convolution network for personalized recommendation of mic ......

Cold Brew: Distilling Graph Node Representations with Incomplete or Missing Neighborhoods

目录概符号说明Cold Brew代码 Zheng W., Huang E. W., Rao N., Katariya S., Wang Z., Subbian K. Cold brew: Distilling graph node representations with incomplete or ......

【五期李伟平】CCF-A(S&P'20)The Value of Collaboration in Convex Machine Learning with Differential Privacy

Nan W., et al. “The Value of Collaboration in Convex Machine Learning with Differential Privacy.” 2020 IEEE Symposium on Security and Privacy. 304-317 ......

LightGCL Simple Yet Effective Graph Contrastive Learning For Recommendation论文阅读笔记

Abstract 目前的图对比学习方法都存在一些问题,它们要么对用户-项目交互图执行随机增强,要么依赖于基于启发式的增强技术(例如用户聚类)来生成对比视图。这些方法都不能很好的保留内在的语义结构,而且很容易受到噪声扰动的影响。所以我们提出了一个图对比学习范式LightGCL来减轻基于CL的推荐者的通 ......

SiReN Sign-Aware Recommendation Using Graph Neural Networks论文阅读笔记

Abstract 目前使用GNN的推荐系统主要利用高评分的正向用户-物品交互信息。但是如何利用低评分来表示用户的偏好是一个挑战,因为低评分仍然可以提供有用的信息。所以在本文中提出了基于GNN模型的有符号感知推荐系统SiReN,SiReN有三个关键组件 构造一个符号二部图更精确的表示用户的偏好,分为两 ......

PANE-GNN Unifying Positive and Negative Edges in Graph Neural Networks for Recommendation论文阅读笔记

Abstract 目前利用GNN的推荐系统主要关注用户的正面反馈,而忽略了负面反馈提供的见解。于是我们提出了PANG- GNN,该模型将图神经网络的正面和负面边统一在一起。PANG-GNN首先将原始评分图根据正面和负面反馈划分为两个不同的二分图。接下来分别使用两个独立的嵌入,即感兴趣嵌入和无兴趣嵌入 ......

Adaptive Graph Contrastive Learning for Recommendation论文阅读笔记

Abstract 在实际的场景中,用户的行为数据往往是有噪声的,并且表现出偏态分布。所以需要利用自监督学习来改善用户表示。我们提出了一种新的自适应图对比学习(AdaGCL)框架,该框架使用两个自适应对比视图生成器来进行数据增强,以更好地增强CF范式。具体的说,我们使用了两个可训练的视图生成器,一个图 ......

Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach

目录概InstructRecInstruction Generation Zhang J., Xie R., Hou Y., Zhao W. X., Lin L., Wen J. Recommendation as instruction following: a large language mo ......

A Novel Approach Based on Bipartite Network Recommendation and KATZ Model to Predict Potential Micro-Disease Associations

A Novel Approach Based on Bipartite Network Recommendation and KATZ Model to Predict Potential Micro-Disease Associations Shiru Li 1, Minzhu Xie 1, Xi ......

Generative-Contrastive Graph Learning for Recommendation论文阅读笔记

Abstract 首先介绍了一下GCL的一些缺点,GCL是通过数据增强来构造对比视图,然后通过最大化对比视图之间的互信息来提供自监督信号。但是目前的数据增强技术都有着一定的缺点 结构增强随机退出节点或边,容易破坏用户项目的内在本质 特征增强对每个节点施加相同的尺度噪声增强,忽略的节点的独特特征 所以 ......

On Manipulating Signals of User-Item Graph A Jacobi Polynomial-based Graph Collaborative Filtering

[TOC] Guo J., Du L, Chen X., Ma X., Fu Q., Han S., Zhang D. and Zhang Y. On manipulating signals of user-item graph: A jacobi polynomial-based graph c ......

TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation

目录概TallRec代码 Bao K., Zhang J., Zhang Y., Wang W., Feng F. and He X. TALLRec: An effective and efficient tuning framework to align large language model ......

Variational Autoencoders for Collaborative Filtering论文阅读笔记

摘要 将VAE扩展到具有隐式反馈的协同过滤,这样能够超越线性因子模型。提出了一个具有多项式条件似然的神经生成模型。目前推荐系统用的比较多的是rank指标,这里本文也说明了为什么多项似然非常适合隐式反馈数据建模。相对于高斯函数和逻辑函数更加接近rank损失 马上提出了一个比较有意思的观点,虽然推荐被认 ......

Linkless Link Prediction via Relational Distillation

目录概符号说明LLP代码 Guo Z., Shiao W., Zhang S., Liu Y., Chawla N. V., Shah N. and Zhao T. Linkless link prediction via relational distillation. ICML, 2023. 概 ......

AutoRec Autoencoders Meet Collaborative Filtering 论文阅读笔记

这个是第一篇将自编码器应用到推荐系统中的论文,也是将深度学习应用到推荐系统中的论文。比较老,主要学习它的思想,对输入的编码与重建。这篇文章提出了基于AutoEncoder的协同过滤方法来解决评分预测问题 我们的目标就是设计一个基于项目或者用户的自动编码器,它可以将每个部分观察到的\(r^u(r^i) ......

Distilling Knowledge from Graph Convolutional Networks

目录概符号说明DistillGCNLocal Structure Preserving代码 Yang Y., Qiu J., Song M., Tao D. and Wang X. Distilling knowledge from graph convolutional networks. CVP ......

Decoupled Knowledge Distillation

目录概符号说明DKD代码 Zhao B., Cui Q., Song R., Qiu Y. and Liang J. Decoupled knowledge distillation. CVPR, 2022. 概 对普通的 KD (Knowledge Distillation) 损失解耦得到 Tar ......
Distillation Decoupled Knowledge

Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time

目录概符号说明Pixie Eksombatchai C., Jindal P., Liu J. Z., Liu Y., Sharma R., Sugnet C., Ulrich M. and Leskovec J. Pixie: A system for recommending 3+ billio ......

! CocoaPods 1.10.0 out of date (1.11.0 is recommended).

[!] Xcode - develop for iOS and macOS (Xcode 14.2) ! CocoaPods 1.10.0 out of date (1.11.0 is recommended). CocoaPods is used to retrieve the iOS and m ......
recommended CocoaPods date out 10

论文阅读:Knowledge Distillation via the Target-aware Transformer

摘要 Knowledge distillation becomes a de facto standard to improve the performance of small neural networks. 知识蒸馏成为提高小型神经网络性能的事实上的标准。 Most of the previo ......

A Lightweight Method for Modeling Confidence in Recommendations with Learned Beta Distributions论文阅读笔记

A Lightweight Method for Modeling Confidence in Recommendations with Learned Beta Distributions论文阅读笔记 摘要 ​ 大多数推荐系统并不提供对其决策信心的指示。因此,他们不区分确定的建议和不确定的建议。现 ......

Personalized Transformer for Explainable Recommendation论文阅读笔记

Personalized Transformer for Explainable Recommendation论文阅读笔记 摘要 ​ 自然语言生成的个性化在大量任务中都起着至关重要的作用。比如可解释的推荐,评审总结和对话系统等。在这些任务中,用户和项目ID是个性化的重要标识符。虽然Transfome ......

8 Innovative BERT Knowledge Distillation Papers That Have Changed The Landscape of NLP

8 Innovative BERT Knowledge Distillation Papers That Have Changed The Landscape of NLP Contemporary state-of-the-art NLP models are difficult to be ut ......

Graph-less Collaborative Filtering

目录概符号说明SimRecPrediction-Level DistillationEmbedding-level DistillationAdaptive Contrastive Regularization总的损失代码 Xia L., Huang C., Shi J. and Xu Y. Gra ......

Unbiased Knowledge Distillation for Recommendation

目录概UnKD代码 Chen G., Chen J., Feng F., Zhou S. and He X. Unbiased knowledge distillation for recommendation. WSDM, 2023. 概 考虑流行度偏差的知识蒸馏, 应用于推荐系统. UnKD M ......

Knowledge Distillation from A Stronger Teacher

目录概DIST代码 Huang T., You S., Wang F., Qian C. and Xu C. Knowledge distillation from a stronger teacher. NIPS, 2022. 概 用 Pearson correlation coefficient ......
Distillation Knowledge Stronger Teacher from
共108篇  :1/4页 首页上一页1下一页尾页