Query：
学者姓名：徐大川
Refining：
Year
Type
Indexed by
Source
Complex
Co-Author
Language
Clean All
Abstract ：
Integrating Knowledge Graphs (KGs) into recommendation systems as supplementary information has become a prevalent strategy. By leveraging the semantic relationships between entities in KGs, recommendation systems can better comprehend user preferences. Due to the unique structure of KGs, methods based on Graph Neural Networks (GNNs) have emerged as the current technical trend. However, existing GNN-based methods struggle to (1) filter out noisy information in real-world KGs, and (2) differentiate the item representations obtained from the knowledge graph and bipartite graph. In this paper, we introduce a novel model called Attention-enhanced and Knowledge-fused Dual item representations Network for recommendation (namely AKDN) that employs attention and gated mechanisms to guide aggregation on both knowledge graphs and bipartite graphs. In particular, we firstly design an attention mechanism to determine the weight of each edge in the information aggregation on KGs, which reduces the influence of noisy information on the items and enables us to obtain more accurate and robust representations of the items. Furthermore, we exploit a gated aggregation mechanism to differentiate collaborative signals and knowledge information, and leverage dual item representations to fuse them together for better capturing user behavior patterns. We conduct extensive experiments on two public datasets which demonstrate the superior performance of our AKDN over state-of-the-art methods, like Knowledge Graph Attention Network (KGAT) and Knowledge Graph- based Intent Network (KGIN).
Keyword ：
Attention mechanisms Attention mechanisms Knowledge engineering Knowledge engineering Collaboration Collaboration Logic gates Logic gates Recommender systems Recommender systems Knowledge Graph (KG) Knowledge Graph (KG) Knowledge graphs Knowledge graphs Graph Neural Networks (GNN) Graph Neural Networks (GNN) Noise measurement Noise measurement Semantics Semantics Market research Market research Bipartite graph Bipartite graph Recommender System (RS) Recommender System (RS) attention mechanism attention mechanism
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Hua, Qiang , Zhou, Jiachao , Zhang, Feng et al. Attention-Enhanced and Knowledge-Fused Dual Item Representations Network for Recommendation [J]. | TSINGHUA SCIENCE AND TECHNOLOGY , 2025 , 30 (2) : 585-599 . |
| MLA | Hua, Qiang et al. "Attention-Enhanced and Knowledge-Fused Dual Item Representations Network for Recommendation" . | TSINGHUA SCIENCE AND TECHNOLOGY 30 . 2 (2025) : 585-599 . |
| APA | Hua, Qiang , Zhou, Jiachao , Zhang, Feng , Dong, Chunru , Xu, Dachuan . Attention-Enhanced and Knowledge-Fused Dual Item Representations Network for Recommendation . | TSINGHUA SCIENCE AND TECHNOLOGY , 2025 , 30 (2) , 585-599 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
An established fact with far-reaching implications in the field of submodular maximization is that the validity of the Jensen-type inequality with probability measures is only necessary but not sufficient for DR-submodular functions. This work aims to demonstrate that a characterization of DR-submodular functions can be achieved by allowing signed measures. We prove this result by leveraging the Lyapunov function approach, which holds independent interest as a general method for deriving inequalities.
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Du,Donglei , Xu,Dachuan . A characterization of DR-submodularity via Jensen-type inequality with signed measure [R]. 2024. (2024-05-07). |
| MLA | Du,Donglei et al. "A characterization of DR-submodularity via Jensen-type inequality with signed measure" 2024. (2024-05-07). |
| APA | Du,Donglei , Xu,Dachuan . A characterization of DR-submodularity via Jensen-type inequality with signed measure 2024. (2024-05-07). |
| Export to | NoteExpress RIS BibTex |
Abstract ：
In this paper, we study online diminishing return submodular (DR-submodular for short) maximization in the bandit setting. Our focus is on problems where the reward functions can be non-monotone, and the constraint set is a general convex set. We first present the Single-sampling Non-monotone Frank-Wolfe algorithm. This algorithm only requires a single call to each reward function, and it computes the stochastic gradient to make it suitable for large-scale settings where full gradient information might not be available. We provide an analysis of the approximation ratio and regret bound of the proposed algorithm. We then propose the Bandit Online Non-monotone Frank-Wolfe algorithm to adjust for problems in the bandit setting, where each reward function returns the function value at a single point. We take advantage of smoothing approximations to reward functions to tackle the challenges posed by the bandit setting. Under mild assumptions, our proposed algorithm can reach 14(1-minx is an element of P '& Vert;x & Vert;infinity)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\frac{1}{4} (1- \min _{x\in {\mathcal {P}}'}\Vert x\Vert _\infty )$$\end{document}-approximation with regret bounded by O(T5min{1,gamma}+56min{1,gamma}+5)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O (T<^>{\frac{5 \min \{1, \gamma \}+5 }{6 \min \{1, \gamma \}+5}})$$\end{document}, where the positive parameter gamma\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma $$\end{document} is related to the "safety domain" P '\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathcal {P}}'$$\end{document}. To the best of our knowledge, this is the first work to address online non-monotone DR-submodular maximization over a general convex set in the bandit setting.
Keyword ：
Non-monotone function Non-monotone function DR submodular maximization DR submodular maximization Bandit setting Bandit setting Regret bound Regret bound Approximation algorithm Approximation algorithm
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Ju, Jiachen , Wang, Xiao , Xu, Dachuan . Online non-monotone diminishing return submodular maximization in the bandit setting [J]. | JOURNAL OF GLOBAL OPTIMIZATION , 2024 , 90 (3) : 619-649 . |
| MLA | Ju, Jiachen et al. "Online non-monotone diminishing return submodular maximization in the bandit setting" . | JOURNAL OF GLOBAL OPTIMIZATION 90 . 3 (2024) : 619-649 . |
| APA | Ju, Jiachen , Wang, Xiao , Xu, Dachuan . Online non-monotone diminishing return submodular maximization in the bandit setting . | JOURNAL OF GLOBAL OPTIMIZATION , 2024 , 90 (3) , 619-649 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
This study investigates electricity trading in a deregulated retail market, aiming to coordinate the energy consumption behavior of end-users through demand response (DR) and carbon emission mechanisms under dynamic pricing. The objective is to minimize integrated energy system (IES) operating cost while maintaining a supply-demand balance. The IES schematic includes generation, dispatch, and consumption, as well as their physical and communication connections, corresponding to the grid, retailer, and multiple PV producers and consumers. Beyond the hierarchical decision-making between the retailer and PV prosumers, the challenge lies in the non-cooperative and competitive interdependence among prosumers. Therefore, the problem is formulated as a Stackelberg-Nash game (SNG) model, with the retailer as leader and PV prosumers as followers. Emphasizing that allowing fair energy trading between prosumers in the Nash game can have a positive impact on pricing decisions. Carbon trading and DR mechanisms enhance the flexibility of coordinating energy system supply-demand balance. An efficient iterative distributed algorithm is proposed for SNG strategy, it can complete real-time scheduling and privacy protection at the same time. Finally, the validity of the model is verified by a case study, which demonstrates that its effectiveness in adapting to changes in energy prices with practicality and economic benefits. Furthermore, we also show that the transactions between prosumers and the DR policies can reduce their operating costs by 12% and 9%, respectively.
Keyword ：
Demand response Demand response Smart grids Smart grids PV prosumers PV prosumers Distributed algorithm Distributed algorithm Stackelberg-Nash game Stackelberg-Nash game
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Yan , Tian, Xiaoyun , Gai, Ling et al. Energy management for PV prosumers inside microgrids based on Stackelberg-Nash game considering demand response [J]. | SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS , 2024 , 68 . |
| MLA | Wu, Yan et al. "Energy management for PV prosumers inside microgrids based on Stackelberg-Nash game considering demand response" . | SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS 68 (2024) . |
| APA | Wu, Yan , Tian, Xiaoyun , Gai, Ling , Lim, Boon-Han , Wu, Tingdong , Xu, Dachuan et al. Energy management for PV prosumers inside microgrids based on Stackelberg-Nash game considering demand response . | SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS , 2024 , 68 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
In this paper, we study approximation algorithms for several classes of DR-submodular optimization problems, where DR is short for diminishing return. Following a newly introduced algorithm framework for zeroth-order stochastic approximation methods, we first propose algorithms CG-ZOSA and RG-ZOSA for smooth DR-submodular optimization based on the coordinate-wise gradient estimator and the randomized gradient estimator, respectively. Our theoretical analysis proves that CG-ZOSA can reach a solution whose expected objective value exceeds (1 e(-1) -is an element of(2))OPT -is an element of after O(-is an element of(-2) ) iterations and O(N-2/3 d is an element of(-2)) oracle calls, where d represents the problem dimension. On the other hand, RG-ZOSA improves the approximation ratio to (1-e(-1)-is an element of(2)/d) while maintaining the same overall oracle complexity. For non-smooth up-concave maximization problems, we propose a novel auxiliary function based on a smoothed objective function and introduce the NZOSA algorithm. This algorithm achieves an approximation ratio of (1-e(-1 )-is an element of ln is an element of(-1)-is an element of(2)ln is an element of(-1)) with O(d is an element of(-2)) iterations and O (N-2/3 d(3/2) is an element of(-3)) oracle calls. We also extend NZOSA to handle a class of robust DR-submodular maximization problems. To validate the effectiveness of our proposed algorithms, we conduct experiments on both synthetic and real-world problems. The results demonstrate the superior performance and efficiency of our methods in solving DR-submodular optimization problems.
Keyword ：
stochastic optimization stochastic optimization DR-submodular optimization DR-submodular optimization zeroth-order gradient estimation zeroth-order gradient estimation approximation algorithm approximation algorithm robust optimization robust optimization
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Lian, Yuefang , Wang, Xiao , Xu, Dachuan et al. Zeroth-order Stochastic Approximation Algorithms for DR-submodular Optimization [J]. | JOURNAL OF MACHINE LEARNING RESEARCH , 2024 , 25 : 1-55 . |
| MLA | Lian, Yuefang et al. "Zeroth-order Stochastic Approximation Algorithms for DR-submodular Optimization" . | JOURNAL OF MACHINE LEARNING RESEARCH 25 (2024) : 1-55 . |
| APA | Lian, Yuefang , Wang, Xiao , Xu, Dachuan , Zhao, Zhongrui . Zeroth-order Stochastic Approximation Algorithms for DR-submodular Optimization . | JOURNAL OF MACHINE LEARNING RESEARCH , 2024 , 25 , 1-55 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Purpose: The formulation and optimization of radiation therapy plans are complex and time-consuming processes that heavily rely on the expertise of medical physicists. Consequently, there is an urgent need for automated optimization methods. Recent advancements in reinforcement learning, particularly deep reinforcement learning (DRL), show great promise for automating radiotherapy planning. This review summarizes the current state of DRL applications in this field, evaluates their effectiveness, and identifies challenges and future directions. Methods: A systematic search was conducted in Google Scholar, PubMed, IEEE Xplore, and Scopus using keywords such as "deep reinforcement learning", "radiation therapy", and "treatment planning". The extracted data were synthesized for an overview and critical analysis. Results: The application of deep reinforcement learning in radiation therapy plan optimization can generally be divided into three categories: optimizing treatment planning parameters, directly optimizing machine parameters, and adaptive radiotherapy. From the perspective of disease sites, DRL has been applied to cervical cancer, prostate cancer, vestibular schwannoma, and lung cancer. Regarding types of radiation therapy, it has been used in HDRBT, IMRT, SBRT, VMAT, GK, and Cyberknife. Conclusions: Deep reinforcement learning technology has played a significant role in advancing the automated optimization of radiation therapy plans. However, there is still a considerable gap before it can be widely applied in clinical settings due to three main reasons: inefficiency, limited methods for quality assessment, and poor interpretability. To address these challenges, significant research opportunities exist in the future, such as constructing evaluators, parallelized training, and exploring continuous action spaces.
Keyword ：
Radiation therapy Radiation therapy Treatment planning Treatment planning Deep reinforcement learning Deep reinforcement learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, Can , Guo, Yuqi , Lin, Xinyan et al. Deep reinforcement learning in radiation therapy planning optimization: A comprehensive review [J]. | PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS , 2024 , 125 . |
| MLA | Li, Can et al. "Deep reinforcement learning in radiation therapy planning optimization: A comprehensive review" . | PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS 125 (2024) . |
| APA | Li, Can , Guo, Yuqi , Lin, Xinyan , Feng, Xuezhen , Xu, Dachuan , Yang, Ruijie . Deep reinforcement learning in radiation therapy planning optimization: A comprehensive review . | PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS , 2024 , 125 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Lower bounded correlation clustering problem (LBCorCP) is a new generalization of the correlation clustering problem (CorCP). In the LBCorCP, we are given an integer L and a complete labelled graph. Each edge in the graph is either positive or negative based on the similarity of its two endpoints. The goal is to find a clustering of the vertices, each cluster contains at least L vertices, so as to minimize the sum of the number of positive cut edges and negative uncut edges. In this paper, we first introduce the LBCorCP and give three algorithms for this problem. The first algorithm is a random algorithm, which is designed for the instances of the LBCorCP with fewer positive edges. The second one is that we let the set V itself as a cluster and prove that the algorithm works well on two specially instances with fewer negative edges. The last one is an LP-rounding based iterative algorithm, which is also provided for the instances with fewer negative edges. The above three algorithms can quickly solve some special instances in polynomial time and obtain a smaller approximation ratio. In addition, we conduct simulations to evaluate the performance of our algorithms.
Keyword ：
Correlation clustering Correlation clustering Lower bounded Lower bounded LP-rounding LP-rounding Approximation algorithm Approximation algorithm Polynomial time Polynomial time
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Ji, Sai , Dong, Yinhong , Du, Donglei et al. Approximation algorithms for the lower bounded correlation clustering problem [J]. | JOURNAL OF COMBINATORIAL OPTIMIZATION , 2023 , 45 (1) . |
| MLA | Ji, Sai et al. "Approximation algorithms for the lower bounded correlation clustering problem" . | JOURNAL OF COMBINATORIAL OPTIMIZATION 45 . 1 (2023) . |
| APA | Ji, Sai , Dong, Yinhong , Du, Donglei , Wang, Dongzhao , Xu, Dachuan . Approximation algorithms for the lower bounded correlation clustering problem . | JOURNAL OF COMBINATORIAL OPTIMIZATION , 2023 , 45 (1) . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Stochastic optimization has experienced significant growth in recent decades, with the increasing prevalence of variance reduction techniques in stochastic optimization algorithms to enhance computational efficiency. In this paper, we introduce two projection-free stochastic approximation algorithms for maximizing diminishing return (DR) submodular functions over convex constraints, building upon the Stochastic Path Integrated Differential EstimatoR (SPIDER) and its variants. Firstly, we present a SPIDER Continuous Greedy (SPIDER-CG) algorithm for the monotone case that guarantees a (1- e(-1))OPT- epsilon approximation after O(epsilon(-1)) iterations and O(epsilon(-2)) stochastic gradient computations under the mean-squared smoothness assumption. For the non-monotone case, we develop a SPIDER Frank-Wolfe (SPIDER-FW) algorithm that guarantees a 1/4 (1- minx (x is an element of C)||x||(infinity))OPT- epsilon approximation withO(epsilon(-1)) iterations and O(epsilon (-2)) stochastic gradient estimates. To address the practical challenge associated with a large number of samples per iteration, we introduce a modified gradient estimator based on SPIDER, leading to a Hybrid SPIDER-FW (Hybrid SPIDER-CG) algorithm, which achieves the same approximation guarantee as SPIDER-FW (SPIDER-CG) algorithm with only O(1) samples per iteration. Numerical experiments on both simulated and real data demonstrate the efficiency of the proposed methods.
Keyword ：
Frank-Wolfe algorithm Frank-Wolfe algorithm DR-Submodular DR-Submodular Gradient estimator Gradient estimator Variance reduction Variance reduction Stochastic algorithm Stochastic algorithm
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Lian, Yuefang , Du, Donglei , Wang, Xiao et al. Stochastic Variance Reduction for DR-Submodular Maximization [J]. | ALGORITHMICA , 2023 , 86 (5) : 1335-1364 . |
| MLA | Lian, Yuefang et al. "Stochastic Variance Reduction for DR-Submodular Maximization" . | ALGORITHMICA 86 . 5 (2023) : 1335-1364 . |
| APA | Lian, Yuefang , Du, Donglei , Wang, Xiao , Xu, Dachuan , Zhou, Yang . Stochastic Variance Reduction for DR-Submodular Maximization . | ALGORITHMICA , 2023 , 86 (5) , 1335-1364 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
In this paper, we study the problem of maximizing a nonmonotone nonnegative one-sided sigma-smooth (OSS) function G(x). For a downwards-closed basis polyhedron constraint, we propose two classes of parallel algorithms under deterministic and random settings. For the deterministic nonmonotone OSS problem, we design the Jump-Start Parallel Frank Wolfe (JSPFW) algorithm and obtain an approximation solution of (e(-1) - e(-2) - O(epsilon)). The JSPFW algorithm has (O(log n/epsilon(2))) adaptive rounds and O(n log n/epsilon(2)) queries about function and its gradient value. For the stochastic OSS problem, we design the Stochastic Parallel Frank Wolfe (SPFW) algorithm and get the (e(-1) - e(-2))(1 - O(epsilon))(1 - o(1)) approximation solution. The SPFW algorithm needs O(n(2)) adaptive rounds and consumes O(n(2)) queries about gradient function value.
Keyword ：
Parallel algorithm Parallel algorithm Frank Wolfe Frank Wolfe OSS function OSS function Down-closed Down-closed
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zhang, Hongxiang , Xu, Dachuan , Zhang, Yapu et al. Parallelized maximization of nonmonotone one-sided sigma-smooth function [J]. | COMPUTERS & ELECTRICAL ENGINEERING , 2023 , 105 . |
| MLA | Zhang, Hongxiang et al. "Parallelized maximization of nonmonotone one-sided sigma-smooth function" . | COMPUTERS & ELECTRICAL ENGINEERING 105 (2023) . |
| APA | Zhang, Hongxiang , Xu, Dachuan , Zhang, Yapu , Zhang, Zhenning . Parallelized maximization of nonmonotone one-sided sigma-smooth function . | COMPUTERS & ELECTRICAL ENGINEERING , 2023 , 105 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
We study a problem called the k-means problem with penalties (k-MPWP), which is a natural generalization of the typical k-means problem. In this problem, we have a set D of client points in R-d, a set F of possible centers in R-d, and a penalty cost p(j) > 0 for each point j is an element of D. We are also given an integer k which is the size of the center point set. We want to find a center point set S subset of F with size k, choose a penalized subset of clients P subset of D, and assign every client in D\P to its open center. Our goal is to minimize the sum of the squared distances between every point in D\P to its assigned centre point and the sum of the penalty costs for all clients in P. By using the multi-swap local search technique and under the fixed-dimensional Euclidean space setting, we present a polynomial-time approximation scheme (PTAS) for the k-MPWP.
Keyword ：
Local search Local search Penalty Penalty k-means k-means Approximation algorithm Approximation algorithm
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yuan, Fan , Xu, Da-Chuan , Du, Dong-Lei et al. Local search yields a PTAS for fixed-dimensional k-means problem with penalties [J]. | JOURNAL OF THE OPERATIONS RESEARCH SOCIETY OF CHINA , 2022 , 12 (2) : 351-362 . |
| MLA | Yuan, Fan et al. "Local search yields a PTAS for fixed-dimensional k-means problem with penalties" . | JOURNAL OF THE OPERATIONS RESEARCH SOCIETY OF CHINA 12 . 2 (2022) : 351-362 . |
| APA | Yuan, Fan , Xu, Da-Chuan , Du, Dong-Lei , Zhang, Dong-Mei . Local search yields a PTAS for fixed-dimensional k-means problem with penalties . | JOURNAL OF THE OPERATIONS RESEARCH SOCIETY OF CHINA , 2022 , 12 (2) , 351-362 . |
| Export to | NoteExpress RIS BibTex |
Export
| Results： |
Selected to |
| Format： |
