{"title":"Deep reinforcement learning-based strategic bidding in electricity markets via variational autoencoder-assisted competitor behavior learning","authors":"Fei Hu, Yong Zhao, Yaowen Yu, Yuanzheng Li","doi":"10.1016/j.engappai.2025.112205","DOIUrl":"10.1016/j.engappai.2025.112205","url":null,"abstract":"<div><div>In a deregulated electricity market, self-interested producers have incentives to offer strategically for maximizing their own profits. While deep reinforcement learning (DRL) has shown great potential for solving such strategic bidding problems, existing methods typically oversimplify strategic action spaces and neglect the influence of competitors' offering behaviors. To bridge these gaps, this paper proposes a novel DRL-based framework to model and solve the strategic bidding problem of an individual producer by jointly considering price-quantity offering actions and the dynamic behaviors of market competitors. First, a bilevel optimization model is formulated to incorporate offering actions on price-quantity pairs. Then, a data-driven framework that combines a variational autoencoder with a density-based clustering method is proposed to learn and capture competitors' offering behaviors. Finally, an imitation learning-integrated DRL algorithm is developed to improve learning stability and solution quality for strategic bidding with price-quantity actions and competitors' offering behaviors. Case studies on the IEEE-30 bus system show that the proposed framework obtains a 28.12 k$ (24.25 %) increase in average profit compared to the existing approach, demonstrating its effectiveness and adaptability under dynamic market conditions.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"161 ","pages":"Article 112205"},"PeriodicalIF":8.0,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deepika Rani Sona, M. Leeban Moses, M. Ramkumar Raja, T. Perarasi
{"title":"Optimized Federated Learning and Blockchain-Based Crowd Sensing for Secure 5G Vehicular Networks","authors":"Deepika Rani Sona, M. Leeban Moses, M. Ramkumar Raja, T. Perarasi","doi":"10.1002/dac.70212","DOIUrl":"https://doi.org/10.1002/dac.70212","url":null,"abstract":"<div>\u0000 \u0000 <p>Intelligent transportation systems (ITS) and the Internet of Vehicles (IoV) face significant challenges in ensuring data security, privacy, and low-latency communication for vehicular crowd sensing. These challenges are exacerbated by rapid node mobility, bursty interactions, and vulnerabilities in existing systems, such as unauthorized data access, delayed message transmission, and inefficient blockchain consensus. To address these challenges, an Optimized Federated Learning and Blockchain-Based Crowd Sensing for Secure 5G Vehicular Networks (OFGNN-BTCS-5G-IoV) is proposed in this paper. Here, the system model is initialized, and the federated generative adversarial network (FGAN) is employed to select relevant active miners and transactions. The FGAN is optimized using the pelican optimization algorithm (POA) to determine optimal parameters to decrease uploading delay. Then the blockchain architecture is used to enhance data storage, transparent validation, and efficient miner selection by addressing privacy and scalability challenges using the Lightweight Proof of Game (LPoG) consensus mechanism. The proposed OFGNN-BTCS-5G-IoV method is implemented in MATLAB, and the OFGNN-BTCS-5G-IoV achieves 20.28%, 28.22%, and 29.27% higher active miner efficiency with 18.26%, 15.22%, and 12.27% lower latency when compared with existing methods. By using FGAN, bio-inspired optimization, and blockchain, the OFGNN-BTCS-5G-IoV offers secure and low-latency vehicular crowd sensing for next-generation ITS.</p>\u0000 </div>","PeriodicalId":13946,"journal":{"name":"International Journal of Communication Systems","volume":"38 15","pages":""},"PeriodicalIF":1.8,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Renwang Song , Chenyu Jiao , Hui Shi , Linying Chen
{"title":"Intelligent compound fault decoupling of rolling bearing based on parallel capsule network","authors":"Renwang Song , Chenyu Jiao , Hui Shi , Linying Chen","doi":"10.1016/j.engappai.2025.112206","DOIUrl":"10.1016/j.engappai.2025.112206","url":null,"abstract":"<div><div>Due to the complex working environment of rolling bearings, various components such as inner ring, outer ring, rolling element and cage in bearing may interact with each other, leading to a compound fault formed by a variety of single fault coupling. Most methods generally regard compound faults as single faults, ignoring the interaction between single faults, which is not conducive to decoupling compound faults into multiple single faults and formulating maintenance plans. Moreover, the capsule network requires stacking multiple capsule layers to enhance performance, which significantly increases model parameters and consumes substantial memory resources. Therefore, a compound fault intelligent decoupling method based on parallel capsule network combining dynamic routing and attention routing is proposed in this study. Firstly, the Omni-Scale block is added to the feature extraction part, which can cover different sizes of receptive fields to enhance the feature extraction ability of the network. Secondly, an attention routing module is proposed, which realize the transmission of information from low-level capsules to high-level capsules by calculating the correlation between the same layers. Finally, the parallel capsule decoupling layer is constructed by using dynamic routing and attention routing. This method is especially suitable for practical engineering scenarios where compound bearing fault samples are limited and computational resources are constrained, providing a lightweight and effective solution for intelligent fault diagnosis. Experimental results show that the proposed method significantly reduces model complexity while maintaining high diagnostic performance under small-sample conditions, with the ablation study further confirming the meaningful contribution of each core module.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"161 ","pages":"Article 112206"},"PeriodicalIF":8.0,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sicong Deng , Jonas Wieskamp , Henrik Born , Heiner Hans Heimes , Achim Kampker
{"title":"Modeling flexible configuration of cell finishing for future battery production research","authors":"Sicong Deng , Jonas Wieskamp , Henrik Born , Heiner Hans Heimes , Achim Kampker","doi":"10.1016/j.rcim.2025.103119","DOIUrl":"10.1016/j.rcim.2025.103119","url":null,"abstract":"<div><div>Today, the increasing demand for battery cells requires efficient large-scale production. At the same time, cell design continues to improve regarding various performance metrics causing product feature changes, which in turn affect the process chain, equipment and process parameter design in production. In cell finishing – the final cell production section – both external cell features, related to the system components, and internal cell features, related to process protocol, are directly affected. However, the interrelations between the core domains – product, process, parameters and equipment – are hardly assessed in the current cell finishing planning and thus no systematic approach to configuration design has been established. This paper focuses on this research need and presents an approach through configuration modeling based on Modularization and Knowledge-Based Design and uses real data from factory planning. From the analyzed raw data, a structured database of product, process, parameters, equipment and their interrelations are derived. For this modeling approach, the paper first explains the conceptual framework. Then, it introduces domains and sub-domains of the database and their formalization for modeling. Subsequently, the architecture of a Two-Stage Configuration Model is explained for flexible configurations (first stage) and virtual modeling (second stage). Finally, the modeling approach is implemented for a real case of prismatic cell finishing. It demonstrates how various configurations can be systematically generated and visualized based on design requirements to advance design optimizations in cell finishing for research purposes and industrial application.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"98 ","pages":"Article 103119"},"PeriodicalIF":11.4,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145007612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An algorithm for sparse piece-wise polynomial residual generation for anomalies detection in industrial manipulator robots","authors":"Mazen Alamir , Sacha Clavel","doi":"10.1016/j.conengprac.2025.106545","DOIUrl":"10.1016/j.conengprac.2025.106545","url":null,"abstract":"<div><div>This paper addresses the characterization of normality in industrial robots which is a crucial step in anomaly detection without an a priori knowledge of the set of faulty behaviors. This is done using piece-wise multi-variate sparse polynomials involved in the identification of implicit relationships between sensors measurements. These relationships are then used to design tight residual generators for complex dynamical systems such as multi-link robot manipulators. An algorithm is provided and tested using real data collected on a 4-link <span>Staübli</span> robot. The precision of the residual is also compared to the one that might be obtained using the state-of-the art Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) showing higher precision with far less complexity and drastically lower computation time. Moreover, some examples of use of the resulting residual generators in anomaly detection are provided showing promising preliminary results.</div></div>","PeriodicalId":50615,"journal":{"name":"Control Engineering Practice","volume":"165 ","pages":"Article 106545"},"PeriodicalIF":4.6,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145003714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"DPMF-Net: A dual-path perceptive multi-stage fusion network for skin lesion segmentation","authors":"Yuling Huang, Yaoyao Ma, Jing Wang, Chao Xu, Zhiwei Fan, Di Wu","doi":"10.1016/j.engappai.2025.112043","DOIUrl":"10.1016/j.engappai.2025.112043","url":null,"abstract":"<div><div>Accurate segmentation of skin lesions in dermoscopic images is crucial for skin cancer detection and treatment. Despite progress in deep learning-based methods, challenges remain due to diverse skin lesion shapes, colors, and blurred boundaries. We propose a novel Dual-path Perceptive Multi-stage Fusion Network (DPMF-Net) for skin lesion segmentation. DPMF-Net integrates multiple feature refinement modules. It aims to gradually optimize lesion representations by leveraging the dual-path framework to perceive high-level contextual information. The Spatial Frequency Dual-path Cascaded Perception Module (SFDCP) synergizes spatial and frequency domains to model long-range dependencies and suppress noise, enhancing perception of low-contrast lesions. Subsequent to the SFDCP, the Spatial Channel Dual-path Parallel Perception Module (SCDPP) employs entropy-driven attention and multi-granularity convolutions in skip connections to dynamically select informative channels and extract lesion details across spatial scales. To verify the efficacy of our proposed DPMF-Net, extensive experimental assessments are carried out across four challenging datasets. The outcomes of these quantitative and qualitative experiments confirm that our approach significantly outperforms current state-of-the-art methods in terms of all evaluation metrics.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"161 ","pages":"Article 112043"},"PeriodicalIF":8.0,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Quality of human-GenAI collaboration and its driving factors: A symbiotic agency perspective","authors":"Jiayu Shang , Dan Huang , Songshan (Sam) Huang","doi":"10.1016/j.ipm.2025.104373","DOIUrl":"10.1016/j.ipm.2025.104373","url":null,"abstract":"<div><div>Generative AI (GenAI) is increasingly integrated into users’ daily work as a collaborative partner. Drawing on the symbiotic agency theory, this study investigates the quality of human-GenAI collaboration using a mixed-methods approach, including qualitative interviews and quantitative surveys. Study 1 identified three dimensions of human-GenAI collaboration quality that comprise outcome quality, comfort, and efficiency; and six driving factors which can be categorized under human agency (domain knowledge, desire for control, and domestication ability), and GenAI agency (communication ability, working memory, and long-term memory). Study 2 applied fuzzy-set qualitative comparative analysis (fsQCA) to explore configurations of the driving factors that lead to high collaboration quality. Four distinct configurations emerged: 1) Domain knowledge, domestication ability, communication ability, and working memory; 2) Domain knowledge, domestication ability, working memory, and long-term memory; 3) Domain knowledge, communication ability, working memory, and long-term memory; and 4) Domain knowledge, ∼desire for control, domestication ability, communication ability, and long-term memory. These results advance understanding of human-GenAI collaboration by highlighting critical configurations of human and GenAI agency that foster high-quality collaboration. The study offers actionable insights to enhance human-GenAI interactions by optimizing both human and GenAI capabilities.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104373"},"PeriodicalIF":6.9,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xiaoyu Fang, Lili Zhang, Haoran Li, Yaowen Zhang, Yunsheng Yao
{"title":"Micro-dynamics prediction of well water level based on GRU and attention mechanism","authors":"Xiaoyu Fang, Lili Zhang, Haoran Li, Yaowen Zhang, Yunsheng Yao","doi":"10.1007/s10489-025-06855-x","DOIUrl":"10.1007/s10489-025-06855-x","url":null,"abstract":"<div><p>Well water level is an important precursor observation, which is expected to be used to extract information on subsurface stress and media changes. Real-time prediction of well water level can help prevent geological disasters, but there are few related experimental studies. This study aims to explore a short-term prediction model of well water level that is more pervasive than the GRU model, explore new methods to enhance the model’s capability, and provide scientific references for the application of deep learning models in the field of well water level prediction. Taking the measured data of the Three Gorges well network from 2012 to 2014 as an example, the performance of the GRU and its variant models on the RMSE, MAE and R² evaluation criteria are compared, and the results show that only the BiGRU-Attention model shows excellent performance at all well points, with better pervasiveness and stability; performing a single-step prediction and adding a 1% standard deviation noise to the training set can improve the robustness and generalisation of the model.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sebastian Clemens Bartsch, Long Hoang Nguyen, Jan-Hendrik Schmidt, Guangyu Du, Martin Adam, Alexander Benlian, Ali Sunyaev
{"title":"The Present and Future of Accountability for AI Systems: A Bibliometric Analysis","authors":"Sebastian Clemens Bartsch, Long Hoang Nguyen, Jan-Hendrik Schmidt, Guangyu Du, Martin Adam, Alexander Benlian, Ali Sunyaev","doi":"10.1007/s10796-025-10636-9","DOIUrl":"https://doi.org/10.1007/s10796-025-10636-9","url":null,"abstract":"<p>Artificial intelligence (AI) systems, particularly generative AI systems, present numerous opportunities for organizations and society. As AI systems become more powerful, ensuring their safe and ethical use necessitates accountability, requiring actors to explain and justify any unintended behavior and outcomes. Recognizing the significance of accountability for AI systems, research from various research disciplines, including information systems (IS), has started investigating the topic. However, accountability for AI systems appears ambiguous across multiple research disciplines. Therefore, we conduct a bibliometric analysis with 5,809 publications to aggregate and synthesize existing research to better understand accountability for AI systems. Our analysis distinguishes IS research, defined by the Web of Science “Computer Science, Information Systems” category, from related non-IS disciplines. This differentiation highlights IS research’s unique socio-technical contribution while ensuring and integrating insights from across the broader academic landscape on accountability for AI systems. Building on these findings, we derive research propositions to lead future research on accountability for AI systems. Finally, we apply these research propositions to the context of generative AI systems and derive a research agenda to guide future research on this emerging topic.</p>","PeriodicalId":13610,"journal":{"name":"Information Systems Frontiers","volume":"31 1","pages":""},"PeriodicalIF":5.9,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145007147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Min Li, Zhifang Qi, Shaobo Deng, Lei Wang, Xiang Yu
{"title":"Adaptive deep shared latent representation enables novel multi-omics cancer subtype classification","authors":"Min Li, Zhifang Qi, Shaobo Deng, Lei Wang, Xiang Yu","doi":"10.1007/s10489-025-06848-w","DOIUrl":"10.1007/s10489-025-06848-w","url":null,"abstract":"<div><p>Variations in outcomes among cancer patients are significant even when they have the same type of tumor. Identifying and classifying molecular subtypes of cancer offers a valuable opportunity to enhance prognosis and tailor treatment plans for individuals. Recent efforts have been made to generate extensive multidimensional genomic data to achieve this potential. However, existing algorithms still face challenges in integrating and analyzing such intricate datasets. In this study, we present Adaptive Deep Shared Latent Representation (ADSLR), a novel approach for cancer subtyping that utilizes shared latent representation to reveal distinct molecular subtypes in cancer. It incorporates a cycle autoencoder with a nonnegative matrix factorization layer, capturing consistent signals of nonlinear features at various omics levels. This enables the generation of adaptable representations for shared latent representation across multiple omics levels. We apply ADSLR to multi-omics data obtained from eight different cancer types in the “The Cancer Genome Atlas” dataset, demonstrating significant improvements in the identification of biologically meaningful cancer subtypes. These identified subtypes exhibit noteworthy variations in patient survival rates across seven out of the eight cancer types. Our analysis uncovers integrated patterns involving mRNA expression, miRNA expression, DNA methylation, and protein across multiple cancers while showcasing ADSLR’s versatility for integrating various other omics types.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144990540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}