Author name cluster

Xiaobing Yu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

1 author row

AAAI Conference 2026 Conference Paper

MoEA-Net: Modality-Incremental Expert Aggregation Network for Retinal Prognostic Prediction

Hua Wang
Xiaodan Zhang
Yanzhao Shi
Chengxin Zheng
Wanyu Zhang
Zhen Wang
Jianing Wang
Xiaobing Yu

Automated analysis of temporal changes in multimodal retinal images is critical for the prognostic assessment of ophthalmic diseases. Unlike traditional single-timepoint diagnosis, tracking longitudinal changes across multiple imaging modalities introduces significant data bias challenges: (1) Imbalanced modality samples compromise the integration of knowledge within minority modalities; (2) Heterogeneous visual patterns across modalities undermine the perception of disease-relevant biomarkers. To tackle these issues, we propose a Modality-Incremental Expert Aggregation Network (MoEA-Net), which unifies the inter-modal integration and intra-modal perception for enhanced retinal prognostic prediction. Specifically, we employ the large language model (LLM) with incremental LoRA layers for specific modalities to effectively integrate knowledge from imbalanced data. Besides, we introduce a Spatiotemporal-aware Expert (SAE) module to better perceive both the anatomical structures and longitudinal changes within modalities. By progressively combining the SAE module with incremental LoRA, MoEA-Net supports continual knowledge accumulation and improves accurate reasoning. Experimental results show that MoEA-Net achieves state-of-the-art performance on subretinal fluid change and visual recovery classification tasks, validating its effectiveness.

PDF Details DOI

EAAI Journal 2025 Journal Article

An improved reinforcement learning-based differential evolution algorithm for combined economic and emission dispatch problems

Yuan Wang
Xiaobing Yu
Wen Zhang

To overcome challenges posed by escalating environmental pollution and climate change, the combined economic and emission dispatch problem is proposed to balance economic efficiency with emission cost. The primary objective of the problem is to ensure that emissions are minimized while optimal economic costs are achieved simultaneously. However, due to the nonlinear and nonconvex characteristics of the model, the optimization is confronted with many difficulties. Hence, an innovative improved reinforcement learning-based differential evolution algorithm is proposed in this article, with reinforcement learning seamlessly integrated into the differential evolution algorithm. Q-learning from reinforcement learning technique is utilized to dynamically adjust parameter settings and select appropriate mutation strategies, thereby boosting the algorithm's adaptability and overall performance. The effectiveness of the proposed algorithm is tested on thirty testing functions and combined economic and emission dispatch problems in comparison with the other five algorithms. According to the experimental results of testing functions, superior performance is consistently achieved by the proposed algorithm, with the highest adaptability exhibited and an average ranking of 1. 4167. Its superiority is further demonstrated through Wilcoxon tests on results of testing functions and combined economic and emission dispatch problems with the proportion of 100%, and the proposed algorithm is significantly better than other algorithms at a 0. 05 significance level. The superiority of the proposed algorithm in optimizing combined economic and emission dispatch problems demonstrates that the proposed algorithm is shown to be adaptable to complex optimization environments, which proves useful for industrial applications and artificial intelligence.

Details DOI

EAAI Journal 2024 Journal Article

Reinforcement learning-based differential evolution algorithm for constrained multi-objective optimization problems

Xiaobing Yu
Pingping Xu
Feng Wang
Xuming Wang

Many real-world problems can be established as Constrained Multi-objective Optimization Problems (CMOPs). It is still challenging to automatically set efficient parameters for Constrained Multi-Objective Evolutionary Algorithms (CMOEAs) to solve these CMOPs. A Reinforcement Learning-based Multi-Objective Differential Evolution (RLMODE) algorithm is proposed, in which the main parameters are dynamically adjusted. During the evolution process, the offspring generated is evaluated and compared with its corresponding parents, the relationship between the offspring and parent can adjust the parameters of RLMODE by the Reinforcement Learning (RL) technique. The feedback mechanism can produce the most appropriate parameters for RLMODE, which pushes the population towards feasible regions. The proposed RLMODE is evaluated on thirty functions and compared with some popular CMOEAs. The performance indicator IGD has revealed that the proposed RLMODE is competitive. Then, they are applied to solve the UAV path planning problem with three objectives and a constraint. The real application has further demonstrated the superiority of the proposed RLMODE.

Details DOI

EAAI Journal 2023 Journal Article

Ranking teaching–learning-based optimization algorithm to estimate the parameters of solar models

Xiaobing Yu
Zhengpeng Hu
Xuming Wang
Wenguan Luo

As one of the most promising renewable energies, solar energy can be converted to electricity through photovoltaic (PV) systems. It is indispensable to identify the parameters of PV systems with the aim of controlling and simulating. Thanks to the complexity of PV systems, parameter identification is still a challenging task. In this paper, we develop a Ranking Teaching–Learning-Based Optimization (RTLBO) to solve the problem, in which Teaching–Learning-Based Optimization (TLBO) is a population-based swarm algorithm and mimics the learning process in a classroom. RTLBO ranks learners into superior and inferior groups, in which the outstanding learners learn from the top three agents to boost the local search. In contrast, the low learners learn from each other by guidance. The two phases are in parallel to balance the local and global search. The proposed RTLBO is used to extract parameters of different models, including the single diode model, double diode model and three PV module models. TLBO, four TLBO variants, and fifteen meta-heuristic algorithms are selected as the rivals of RTLBO. Several experiments have shown that our method is a reliable and effective algorithm when addressing the parameters of PV systems.

Details DOI

EAAI Journal 2023 Journal Article

Solving combined economic and emission dispatch problems using reinforcement learning-based adaptive differential evolution algorithm

Wenguan Luo
Xiaobing Yu
Yifan Wei

Nowadays, economic and environmental concerns in production have become increasingly significant. To address these issues, the Combined Economic and Emission Dispatch (CEED) problem has been introduced to optimize the power generation process by considering fuel cost and emitted substances. However, due to the nonlinearity and nonconvexity of the objective function, the optimization of CEED remains a challenge. In this paper, we develop a Reinforcement Learning-based Adaptive Differential Evolution (RLADE) algorithm to enhance the optimization performance. The mutation strategy and crossover probability of RLADE are optimized using Reinforcement Learning (RL) to respectively ensure better convergence speed and searchability. Additionally, two modifications of RL, namely the adaptive population size-based state division and fitness-ranking-based reward mechanism, are proposed to improve the accuracy of state division and reward calculation in RL. The experiments conducted in this paper consider two objective formulation methods of CEED problems, namely the quadratic and cubic criterion functions. The mean values and standard deviations of the obtained solutions were utilized to assess the performance of RLADE, as well as other comparative algorithms, namely DE algorithm and two RL-based DE variants. The results clearly demonstrate that RLADE surpasses its counterparts with proportion of 100%, 85. 7%, and 100% for the 6-unit and 11-unit quadratic CEED problems, as well as cubic criterion functions, in terms of both search accuracy and convergence ability. Furthermore, the significance of RLADE's superiority is confirmed through the Wilcoxon's signed rank test.

Details DOI

EAAI Journal 2022 Journal Article

Consensus reaching model for counter-intuitive in D–S evidence theory and application under 2-tuple linguistic representation

Chenliang Li
Xiaobing Yu

In the information fusion field, Dempster–Shafer (D–S) evidence theory is a multi-source technology to solve the uncertain problems. With the aim of improving the decision accuracy, D–S evidence theory can make full use of information from different sources which are redundant and complementary. However, the influence caused by conflicting evidence during information fusion, named the counter-intuitive result, will confuse the selection of decision-makers (DMs). Inspired by the distance-based uncertainty measure, which is a typical technique used to manage uncertain information, the consensus reaching model is applied to overcome the influence caused by conflicting evidence in this paper. Considering the hesitant and uncertain of cognition, 2-tuple linguistic representation method is introduced to model and manage this vague decision information given by DMs. Finally, a consensus reaching model for counter-intuitive result in D–S evidence theory is put forward. To verify the effectiveness of the proposed method, the selection of plant protection machine suppliers is modeled as a multi-criteria decision-making (MCDM) problem. According to the decision-making process, the best option for plant protection machine suppliers is obtained. The decision result indicates a strong correlation between the consistency value and conflicting value of evidence.

Details DOI