Pub Date : 2026-02-09DOI: 10.1109/tnnls.2026.3657138
Basit Alawode, Iyyakutti Iyappan Ganapathi, Sajid Javed, Mohammed Bennamoun, Arif Mahmood
{"title":"AquaticCLIP: A Vision-Language Foundation Model and Dataset for Underwater Scene Analysis","authors":"Basit Alawode, Iyyakutti Iyappan Ganapathi, Sajid Javed, Mohammed Bennamoun, Arif Mahmood","doi":"10.1109/tnnls.2026.3657138","DOIUrl":"https://doi.org/10.1109/tnnls.2026.3657138","url":null,"abstract":"","PeriodicalId":13303,"journal":{"name":"IEEE transactions on neural networks and learning systems","volume":"161 1","pages":""},"PeriodicalIF":10.4,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146146044","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}
Pub Date : 2026-02-09DOI: 10.1109/jiot.2026.3662758
Haowen Zhang, Juan Li, Qing Yao
{"title":"RACER: Fast and Accurate Time Series Clustering with Random Convolutional Kernels and Ensemble Methods","authors":"Haowen Zhang, Juan Li, Qing Yao","doi":"10.1109/jiot.2026.3662758","DOIUrl":"https://doi.org/10.1109/jiot.2026.3662758","url":null,"abstract":"","PeriodicalId":54347,"journal":{"name":"IEEE Internet of Things Journal","volume":"314 1","pages":""},"PeriodicalIF":10.6,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146146092","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}
Pub Date : 2026-02-09DOI: 10.1007/s10796-026-10698-3
Mustafa Cavus, Jan N. van Rijn, Przemysław Biecek
Trustworthiness of AI systems is a core objective of Human-Centered Explainable AI, and relies, among other things, on explainability and understandability of the outcome. While automated machine learning tools automate model training, they often generate not only a single “best” model but also a set of near-equivalent alternatives, known as the Rashomon set. This set provides a unique opportunity for human-centered explainability: by exposing variability among similarly performing models, we can offer users richer and more informative explanations. In this paper, we introduce Rashomon partial dependence profiles , a model-agnostic technique that aggregates feature effect estimates across the Rashomon set. Unlike traditional explanations derived from a single model, Rashomon partial dependence profiles explicitly quantify uncertainty and visualize variability, further enabling user trust and understanding model behavior to make informed decisions. Additionally, under high-noise conditions, the Rashomon partial dependence profiles more accurately recover ground-truth feature relationships than a single-model partial dependence profile. Experiments on synthetic and real-world datasets demonstrate that Rashomon partial dependence profiles reduce average deviation from the ground truth by up to 38%, and their confidence intervals reliably capture true feature effects. These results highlight how leveraging the Rashomon set can enhance technical rigor while centering explanations on user trust and understanding aligned with Human-centered explainable AI principles.
{"title":"Quantifying Model Uncertainty with AutoML and Rashomon Partial Dependence Profiles: Enabling Trustworthy and Human-centered XAI","authors":"Mustafa Cavus, Jan N. van Rijn, Przemysław Biecek","doi":"10.1007/s10796-026-10698-3","DOIUrl":"https://doi.org/10.1007/s10796-026-10698-3","url":null,"abstract":"Trustworthiness of AI systems is a core objective of Human-Centered Explainable AI, and relies, among other things, on explainability and understandability of the outcome. While automated machine learning tools automate model training, they often generate not only a single “best” model but also a set of near-equivalent alternatives, known as the Rashomon set. This set provides a unique opportunity for human-centered explainability: by exposing variability among similarly performing models, we can offer users richer and more informative explanations. In this paper, we introduce <jats:italic>Rashomon partial dependence profiles</jats:italic> , a model-agnostic technique that aggregates feature effect estimates across the Rashomon set. Unlike traditional explanations derived from a single model, Rashomon partial dependence profiles explicitly quantify uncertainty and visualize variability, further enabling user trust and understanding model behavior to make informed decisions. Additionally, under high-noise conditions, the Rashomon partial dependence profiles more accurately recover ground-truth feature relationships than a single-model partial dependence profile. Experiments on synthetic and real-world datasets demonstrate that Rashomon partial dependence profiles reduce average deviation from the ground truth by up to 38%, and their confidence intervals reliably capture true feature effects. These results highlight how leveraging the Rashomon set can enhance technical rigor while centering explanations on user trust and understanding aligned with Human-centered explainable AI principles.","PeriodicalId":13610,"journal":{"name":"Information Systems Frontiers","volume":"45 1","pages":""},"PeriodicalIF":5.9,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146146038","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}
Pub Date : 2026-02-09DOI: 10.1109/tac.2026.3663109
Pietro A. Refosco, Christopher Edwards, Dimitrios Papageorgiou
{"title":"Conditions for boundedness of under-tuned super-twisting sliding mode control loops","authors":"Pietro A. Refosco, Christopher Edwards, Dimitrios Papageorgiou","doi":"10.1109/tac.2026.3663109","DOIUrl":"https://doi.org/10.1109/tac.2026.3663109","url":null,"abstract":"","PeriodicalId":13201,"journal":{"name":"IEEE Transactions on Automatic Control","volume":"314 1","pages":""},"PeriodicalIF":6.8,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146146130","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}
Pub Date : 2026-02-09DOI: 10.1016/j.eswa.2026.131603
Yonghong Liu , Ziming Wang , Yupeng Xie , De Huang
Maintaining stable airflow within mine ventilation systems is essential for ensuring safe and continuous underground operations. However, unsteady airflow disturbances induced by the intermittent movement of mine cars or hoisting cages generate transient, low-amplitude perturbations that are dynamically coupled across the ventilation network. These disturbances are superimposed on the steady mechanical ventilation field, producing unsteady signals that conventional steady-state models cannot effectively decouple or localize, leading to discrepancies between monitoring data and actual ventilation conditions. To address this challenge, a mathematical model was developed to characterize unsteady airflow disturbances in underground tunnels, and the dynamic effects of mine car movement on ventilation airflow were systematically analyzed. A network-based algorithm was further designed to solve the unsteady disturbance field, and a simulation platform was constructed to reproduce dynamic airflow behavior, showing minimal deviation from theoretical predictions. Building on this foundation, a hybrid Maximum Information Coefficient-Long Short-Term Memory (MIC-LSTM) neural network model was proposed for the inverse identification of unsteady disturbance sources. The Maximum Information Coefficient (MIC) was utilized to extract informative features from airflow velocity time-series data, while the LSTM network identified disturbance sources from temporal dependencies. Experimental results demonstrate that when the disturbance threshold is 0.1 and the monitoring coverage ratio is 0.3, all evaluation metrics approximately 90%. Validation in an operational mine ventilation system further confirms the model’s accuracy, robustness, and generalizability. This study establishes an artificial intelligence-driven framework for intelligent monitoring and control of unsteady disturbances, providing actionable insights toward safer and more efficient mine ventilation management.
{"title":"Inverse identification of unsteady disturbance sources in mine ventilation systems","authors":"Yonghong Liu , Ziming Wang , Yupeng Xie , De Huang","doi":"10.1016/j.eswa.2026.131603","DOIUrl":"10.1016/j.eswa.2026.131603","url":null,"abstract":"<div><div>Maintaining stable airflow within mine ventilation systems is essential for ensuring safe and continuous underground operations. However, unsteady airflow disturbances induced by the intermittent movement of mine cars or hoisting cages generate transient, low-amplitude perturbations that are dynamically coupled across the ventilation network. These disturbances are superimposed on the steady mechanical ventilation field, producing unsteady signals that conventional steady-state models cannot effectively decouple or localize, leading to discrepancies between monitoring data and actual ventilation conditions. To address this challenge, a mathematical model was developed to characterize unsteady airflow disturbances in underground tunnels, and the dynamic effects of mine car movement on ventilation airflow were systematically analyzed. A network-based algorithm was further designed to solve the unsteady disturbance field, and a simulation platform was constructed to reproduce dynamic airflow behavior, showing minimal deviation from theoretical predictions. Building on this foundation, a hybrid Maximum Information Coefficient-Long Short-Term Memory (MIC-LSTM) neural network model was proposed for the inverse identification of unsteady disturbance sources. The Maximum Information Coefficient (MIC) was utilized to extract informative features from airflow velocity time-series data, while the LSTM network identified disturbance sources from temporal dependencies. Experimental results demonstrate that when the disturbance threshold is 0.1 and the monitoring coverage ratio is 0.3, all evaluation metrics approximately 90%. Validation in an operational mine ventilation system further confirms the model’s accuracy, robustness, and generalizability. This study establishes an artificial intelligence-driven framework for intelligent monitoring and control of unsteady disturbances, providing actionable insights toward safer and more efficient mine ventilation management.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 131603"},"PeriodicalIF":7.5,"publicationDate":"2026-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154281","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：