Yong Deng’s research while affiliated with University of Electronic Science and Technology of China and other places

... Similar to entropy methods as uncertainty measure in evidence theory, RPS entropy has been proposed recently (Chen and Deng, 2023). What is more, the maximum RPS entropy and its limit form are also introduced and proved (Zhan et al., 2024 andZhou et al., 2024). ...

Reference:

Random walk in random permutation set theory

... This theory enables the combination of evidence from different sources and the determination of a degree of belief, represented by a mathematical object called belief functions, that takes into account all available evidence. DST has a wide range of applications in various fields such as multi-source information fusion [3,4], risk assessment [5,6], information entropy [7,8], decision making [9,10], random work simulation [11] and real-world applications such as classification [12], computer vision [13,14] and satellite communications [15,16]. In [17], Smets standardizes the matrix notation for calculating belief functions, where beliefs are represented by one-hot encoding vectors. ...

Reference:

Towards an efficient implementation of Dempster–Shafer: α\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha $$\end{document}-junction fusion rules on quantum circuits

... The part of this paper is an extended version of[45]. ...

Reference:

Transferable Belief Model on Quantum Circuits

... Managing these uncertainties and inconsistencies is a persistent problem. To overcome this issue, researchers have proposed a variety of solutions, including random permutation set (Deng 2022), network-based models (Song and Xiao 2022;Wen and Cheong 2021;, evidence-based strategies (Xu and Qian 217;Song et al. 2022), entropy-based techniques (Cui et al. 2022;Zhou et al. 2021;Babajanyan et al. 2020), and logic-driven frameworks like rough set theory Fei and Feng 2021). ...

Reference:

Boosting complex evidence theory with complex belief Renyi divergence for multi-source information fusion

... Building upon these contributions, this paper's CDM-PSL advances the field by optimizing the balance between solution convergence and diversity through the integration of conditional and unconditional diffusion models. Moreover, CDM-PSL incorporates gradient information, weighted by information entropy (Zhan et al. 2024), into the process of generating solutions, significantly enhancing convergence performance during the early-stage iterations. The combination of these strategies makes CDM-PSL have competitive performance in solving EMOPs. ...

Reference:

Expensive Multi-Objective Bayesian Optimization Based on Diffusion Models

... In the realm of the generalizable, the TBM serves as a bridge between probabilistic and possibilistic information [47], enabling belief functions to model both statistical data and linguistic knowledge. Therefore, the belief function theory, developed from the TBM semantics is widely used in multi-source information fusion [38,15,43], expert decision making [41,6], fault diagnosis [4,17], classifier fusion [1,40] and computer vision [14,13]. However, these approaches typically represent data and knowledge as belief functions using general machine learning methods that perform evidential operations on high-level information representations or small-scale data sets. ...

Reference:

Transferable Belief Model on Quantum Circuits

... In [19], trust risks in the conflict-eliminating process (CEP) of SNGDM were addressed through third-party monitoring, and a trust risk analysis-based conflict-eliminating model for SNGDM was developed. In [20], an interaction indicator was exploited to represent peer interaction effort (PIE), individual social cooperation networks (ISCNs) were constructed with the log-sigmoid transition technique, and a novel game-theoretic expert importance evaluation model guided by cooperation effects was finally proposed. To address the challenge that decision-makers in a community setting usually exhibited complex social preferences and intricate social interactions, a minimum cost consensus-based SNGDM approach considering altruism-fairness preferences and ordered trust propagation was designed in [21]. ...

Reference:

Fuzzy Information Evolution with Three-Way Decision in Social Network Group Decision-Making

... In order to measure the uncertainty of evidence, Deng proposed Deng entropy and information volume of mass function [10], [11]. Deng entropy and information volume of mass function contributes to nonlinear systems [12]- [14] and information measurement [15], also plays a role in time series analysis [16]. Considering quantum theory, Pan et al. proposed quantum combination rules [17]. ...

Reference:

Random Graph Set and Evidence Pattern Reasoning Model

... The first category mainly utilizes the entropy weight method to weigh the evidence to be fused. Specifically, in evidence theory, different pieces of evidence (or information sources) may contribute differently to the decision-making process [50][51][52]. The entropy weight method evaluates the importance or reliability of each piece of evidence by calculating its entropy value. ...

Reference:

A Novel Method for Pignistic Information Fusion in the View of Z-number

... Additionally, for a more flexible linguistic representation, FLPRs based on consensus reaching models should be developed when the sum of proportions is not unity and one or more linguistic terms are adopted. (2) There may exist social network relationships [42,54], complex group opinion fusion [10,55], and conflict analysis [2,5] in GCR problems. Therefore, extending the proposed method to these complex decision situations is a promising research direction. ...

Reference:

A group consensus reaching model balancing individual satisfaction and group fairness for distributed linguistic preference relations