CDARL: a contrastive discriminator-augmented reinforcement learning framework for sequential recommendations

Liu, Zhuang; Ma, Yunpu; Hildebrandt, Marcel; Ouyang, Yuanxin; Xiong, Zhang

doi:10.1007/s10115-022-01711-7

Cited by 7 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an important branch of deep learning, reinforcement learning (RL) has achieved remarkable results in the exploration of sequential recommendation. Many methods based on reinforcement learning achieve more accurate personalized recommendations by designing appropriate reward functions and strategies, such as [25], [26], [27]. Our method differs from RL-based methods in that we concentrate on the extraction of intricate relationships between items in the session, while RL concentrates on the interaction between the agent and the environment.…”

Section: B Deep Neural Network Based Methodsmentioning

confidence: 99%

Intention-Centric Learning via Dual Attention for Sequential Recommendation

Zhang,

Wang,

Xie

2024

IEEE Access

View full text Add to dashboard Cite

In sequential recommendation, it is critical to accurately capture the user's intention with limited session information. Previous work concentrates on modeling a single relationship existing between items in an ongoing session, e.g., sequential dependency or graph structures. They lack the exploration of constituent semantic properties of user intention and also lack effective mechanisms binding important features to deduce user intention accurately. In this paper, we present a novel intention detection-enhanced sequential recommendation model, named DASR, which can capture both sequential dependencies and user intent components. We innovatively introduce slot attention to bind the low-level local features extracted by CNN and design multiple rounds of competitive iteration mechanisms to refine the high-level representation of user intent continuously. Finally, these high-level features cooperate with the global dependencies captured by self-attention to achieve sequential recommendations. Extensive experiments are carried out on three benchmark datasets, the experimental results show that DASR outperforms the state-of-the-art baseline methods up to 6.91%, and 4.73% on Recall@20, and MRR@20, respectively.

show abstract

Section: B Deep Neural Network Based Methodsmentioning

confidence: 99%

Intention-Centric Learning via Dual Attention for Sequential Recommendation

Zhang,

Wang,

Xie

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Recommendation technology emerged to solve the problem of information overload, and personalized recommendation algorithms are computer algorithms used to provide recommendations based on the user's interests, preferences, and history, which can mine the information needed by the user from huge data and then recommend it to the user [2]. The emergence of recommendation algorithms is widely used in the Internet and also in commercial applications, such as short videos, e-commerce, social and other fields [3]. The use of personalized recommendation algorithms in applications in these areas provides users with personalized recommendations that help them discover new content, increase user satisfaction and also attract user retention.…”

Section: Introductionmentioning

confidence: 99%