5,782 Hits in 5.3 sec

User behavior understanding in real world settings [article]

Weiqi Shao, Xu Chen, Jiashu Zhao, Long Xia, Dawei Yin
2022 arXiv   pre-print
The IRC module learns the overall sequential characteristics of user behavior with a bi-directional architecture transformer.  ...  In this paper we propose a model that can automatically and adaptively generates a dynamic group of representations from the user behavior accordingly.  ...  In a summary, the main contributions of this paper can be concluded as follows: • We proposed to build sequential recommender models by adaptively disentangling user preferences, which, to the best of  ... 
arXiv:2112.02812v3 fatcat:ai6q2onxbzar7kd2fqdy5spmia

Disentangled Graph Neural Networks for Session-based Recommendation [article]

Ansong Li, Zhiyong Cheng, Fan Liu, Zan Gao, Weili Guan, Yuxin Peng
2022 arXiv   pre-print
To address the problem, we propose a novel method called Disentangled Graph Neural Network (Disen-GNN) to capture the session purpose with the consideration of factor-level attention on each item.  ...  Session-based recommendation (SBR) has drawn increasingly research attention in recent years, due to its great practical value by only exploiting the limited user behavior history in the current session  ...  For example, the FPMC [7] method, which combines matrix factorization and the first-order Markov chain to model the sequential pattern and user preference for recommendation, can be adapted for SBR by  ... 
arXiv:2201.03482v2 fatcat:3r5k4eh62re2jgfgi27jje76r4

Controllable Recommenders using Deep Generative Models and Disentanglement [article]

Samarth Bhargav, Evangelos Kanoulas
2021 arXiv   pre-print
In this paper, we consider controllability as a means to satisfy dynamic preferences of users, enabling them to control recommendations such that their current preference is met.  ...  We propose an alternate view, where instead of keyphrase based critiques, a user is provided 'knobs' in a disentangled latent space, with each knob corresponding to an item aspect.  ...  [26] apply disentanglement to the sequential recommendation task, while Wang et al. [34] disentangle diverse user-intents using graph based collaborative filtering; Cui et al.  ... 
arXiv:2110.05056v1 fatcat:eepjuxveebffdb2i7vglqx65ra

Sequential Recommendation with Decomposed Item Feature Routing

Kun Lin, Zhenlei Wang, Shiqi Shen, Zhipeng Wang, Bo Chen, Xu Chen
2022 Proceedings of the ACM Web Conference 2022  
Sequential recommendation basically aims to capture user evolving preference.  ...  Intuitively, a user interacts with an item usually because of some specific feature, and user evolving preference is essentially determined by a series of important features along the time line.  ...  These results agree with the previous work [11, 16, 25] , and manifest that modeling user sequential preference is indeed effective for improving the recommendation performance.  ... 
doi:10.1145/3485447.3512101 fatcat:54pm7fq34zcv7ehrl5f6mmn4ca

Disentangled Item Representation for Recommender Systems [article]

Zeyu Cui, Feng Yu, Shu Wu, Qiang Liu, Liang Wang
2021 arXiv   pre-print
To this end, we propose a fine-grained Disentangled Item Representation (DIR) for recommender systems in this paper, where the items are represented as several separated attribute vectors instead of a  ...  Some studies use the given attribute information as side information, which is concatenated with the item latent vector to augment representations.  ...  Since then, some works try to apply BPR to sequential recommendation.  ... 
arXiv:2008.07178v2 fatcat:ljncmzxq6bhd3g3gk62i66oc6y

Intent Disentanglement and Feature Self-supervision for Novel Recommendation [article]

Tieyun Qian, Yile Liang, Qing Li, Xuan Ma, Ke Sun, Zhiyong Peng
2021 arXiv   pre-print
preference).  ...  We then present a unified end-to-end framework to simultaneously optimize accuracy and novelty targets based on the disentangled intent of popularity and that of preference.  ...  [31] propose to disentangle the intents behind any given sequence of behaviors for sequential recommendation.  ... 
arXiv:2106.14388v1 fatcat:m4i73l45fzgzfhp7k3ud4aqc5y

Transition Information Enhanced Disentangled Graph Neural Networks for Session-based Recommendation [article]

Ansong Li
2022 arXiv   pre-print
Session-based recommendation is a practical recommendation task that predicts the next item based on an anonymous behavior sequence, and its performance relies heavily on the transition information between  ...  However, most existing methods treat neighbors from different sessions equally without considering that the neighbor items from different sessions may share similar features with the target item on different  ...  preference.  ... 
arXiv:2204.02119v1 fatcat:erqail6gdjgifnxngrwmtxwvsu

Multi-Behavior Sequential Recommendation with Temporal Graph Transformer

Lianghao Xia, Chao Huang, Yong Xu, Jian Pei
2022 IEEE Transactions on Knowledge and Data Engineering  
Modeling time-evolving preferences of users with their sequential item interactions, has attracted increasing attention in many online applications.  ...  Hence, sequential recommender systems have been developed to learn the dynamic user interests from the historical interactions for suggesting items.  ...  for modeling user dynamic preference in sequential recommendation. • NARM [23] .  ... 
doi:10.1109/tkde.2022.3175094 fatcat:iqreqptfvbeeffmit4isv7xsuu

Modality Matches Modality: Pretraining Modality-Disentangled Item Representations for Recommendation

Tengyue Han, Pengfei Wang, Shaozhang Niu, Chenliang Li
2022 Proceedings of the ACM Web Conference 2022  
CCS CONCEPTS • Information systems → Recommender systems.  ...  To this end, we propose a pretraining framework PAMD, which stands for PretrAining Modality-Disentangled Representations Model.  ...  The parameters Θ of PAMD are further fine-tuned during this recommendation learning phase. Given a user and an item, we compute the preference score with the inner product.  ... 
doi:10.1145/3485447.3512079 fatcat:aburalirkvam3mcyog3cidyoqm

DACSR: Decoupled-Aggregated End-to-End Calibrated Sequential Recommendation [article]

Jiayi Chen, Wen Wu, Liye Shi, Yu Ji, Wenxin Hu, Wei Zheng, Liang He
2022 arXiv   pre-print
On the one hand, it aims to provide fairer recommendations whose preference distributions are consistent with users' historical behaviors.  ...  Sequential recommendations have made great strides in accurately predicting the future behavior of users.  ...  By applying the calibration loss function, the sequential recommendation models became aware of the preference distribution of recommended items, and aligned it to the historical preference distribution  ... 
arXiv:2204.10796v12 fatcat:lkcnsen6qzdq3ku6ebjf3j4apu

Seq2seq Translation Model for Sequential Recommendation [article]

Ke Sun, Tieyun Qian
2020 arXiv   pre-print
The context information such as product category plays a critical role in sequential recommendation. Recent years have witnessed a growing interest in context-aware sequential recommender systems.  ...  In this paper, we propose a novel seq2seq translation architecture to highlight the importance of sequential dependency in contexts for sequential recommendation.  ...  recommendation, users' general long term preference is also important for next item prediction in addition to sequential patterns.  ... 
arXiv:1912.07274v2 fatcat:k4kmbpgpfvbexo375winub5ega

Multi-Interest-Aware User Modeling for Large-Scale Sequential Recommendations [article]

Jianxun Lian, Iyad Batal, Zheng Liu, Akshay Soni, Eun Yong Kang, Yajun Wang, Xing Xie
2021 arXiv   pre-print
Precise user modeling is critical for online personalized recommendation services.  ...  This paper proposes a Sequential User Matrix (SUM) to accurately and efficiently capture users' diverse interests.  ...  The union of the two types of channels empowers the SUM model with the flexibility of switching between interest mixture and interest disentanglement adaptively with different datasets.  ... 
arXiv:2102.09211v3 fatcat:ko5u6yavh5h2ni4isnanhnieue

ADER: Adaptively Distilled Exemplar Replay Towards Continual Learning for Session-based Recommendation [article]

Fei Mi, Xiaoyu Lin, Boi Faltings
2020 arXiv   pre-print
To deal with this challenge, we propose a method called Adaptively Distilled Exemplar Replay (ADER) by periodically replaying previous training samples (i.e., exemplars) to the current model with an adaptive  ...  A major challenge for continual learning with neural models is catastrophic forgetting, in which a continually trained model forgets user preference patterns it has learned before.  ...  To prevent the recommender forgetting user preferences learned before, we propose ADER by replaying carefully chosen exemplars from previous cycles and an adaptive distillation loss.  ... 
arXiv:2007.12000v1 fatcat:6s3dc4exrffk3aacj7dtd5mhfi

Graph Neural Networks in Recommender Systems: A Survey

Shiwen Wu, Fei Sun, Wentao Zhang, Xu Xie, Bin Cui
2022 ACM Computing Surveys  
With the explosive growth of online information, recommender systems play a key role to alleviate such information overload.  ...  We collect the representative papers along with their open-source implementations in  ...  ., how to disentangle the embedding pertinent to users' intents; how to set the diferent interest number for each user in an adaptive way; how to design an eicient and efective propagation schema for multiple  ... 
doi:10.1145/3535101 fatcat:hgv2tbx3k5hzbnkupwsysqwjmy

Privacy-preserving Voice Analysis via Disentangled Representations [article]

Ranya Aloufi, Hamed Haddadi, David Boyle
2020 arXiv   pre-print
Based on a user's preferences, a supervision signal informs the filtering out of invariant factors while retaining the factors reflected in the selected preference.  ...  We leverage disentangled representation learning to explicitly learn independent factors in the raw data.  ...  The model was trained using the CTC loss function and with a Stochastic Gradient Descent (SGD) and Momentum optimizer which was extended with the Layer-wise Adaptive Rate Clipping (LARC) algorithm.  ... 
arXiv:2007.15064v1 fatcat:oancnsvxlja4zdz2whxef5s3tm
« Previous Showing results 1 — 15 out of 5,782 results