Artificial Intelligence Review最新文献

{"title":"Deconstructing deep imbalanced regression: a comprehensive review and experimental evaluation","authors":"Noah C. Puetz,&nbsp;Jens U. Brandt,&nbsp;Marc Hilbert,&nbsp;Elena Raponi,&nbsp;Thomas Bäck,&nbsp;Thomas Bartz-Beielstein","doi":"10.1007/s10462-026-11570-1","DOIUrl":"10.1007/s10462-026-11570-1","url":null,"abstract":"<div><p>In real-world applications, there is a fundamental problem: the data most critical to predict interesting events, anomalies, and high-stakes outliers are the rarest, while less interesting data is abundant. Although deep learning is deployed specifically for these difficult prediction tasks, data-driven models inevitably fail in underrepresented areas. This discrepancy between the empirical data- and the desired evaluation distribution is equivalent to a target distribution shift. The research field, termed Deep Imbalanced Regression (DIR), has emerged explicitly to address this challenge, which is particularly acute for continuous targets where most conventional classification-based methods are ill-suited. In this paper, we present the first comprehensive review of the DIR landscape, organized around a novel two-axis taxonomy that disentangles challenges along a <i>Data Axis</i> (target distribution shift, continuity, and density) and a <i>Deep-Learning Axis</i> (shared capacity, biased updates, and manifold distortion), where the latter captures a cascading failure mechanism through which deep models systematically neglect underrepresented targets. Within this framework, we systematically categorize and analyze 19 state-of-the-art methods spanning architectural, algorithm-level, and representation learning approaches, and empirically re-evaluate twelve of them with publicly available implementations under controlled, identical conditions. To stress-test generalization across the full target range, we introduce three novel targeted evaluation protocols, <i>Balanced Extrapolation</i>, <i>Bimodal Interpolation</i>, and <i>Blind-Spot Isolation</i>, that expose failure modes hidden by standard benchmarks (https://github.com/noah-puetz/deconstructing_deep_imbalanced_regression). Our study underscores the significant impact of imbalance on regression accuracy, offering a conceptual framework and practical benchmarks to catalyze further development of systems capable of capturing the rare as reliably as the common.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 6","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11570-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147754380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trust as a design principle in human–robot collaboration: a review of explainable and adaptive control","authors":"Abdelrahman T. Elgohr,&nbsp;Maher Rashad,&nbsp;Eman M. El-Gendy,&nbsp;Waleed Shaaban,&nbsp;Mahmoud M. Saafan","doi":"10.1007/s10462-026-11554-1","DOIUrl":"10.1007/s10462-026-11554-1","url":null,"abstract":"<div>\u0000 \u0000 <p>Human–Robot Collaboration (HRC) has emerged as a fundamental element of new industrial and service systems, wherein humans and robots function within common physical and cognitive environments to achieve shared goals. In addition to traditional issues of safety and productivity, trust has become a critical element influencing cooperation efficiency, human dependence, and sustained system acceptability. This review offers a thorough and reliability-focused summary of HRC research, highlighting the significance of explainable intelligence and adaptive control in promoting trustworthy collaboration. The study initially identifies trust as a fundamental design target in HRC, delineating its dynamic, multifaceted characteristics and its impact on human decision-making and interaction behavior. A systematic review methodology is employed to analyze cutting-edge approaches across essential dimensions, including trust modeling and estimation, multimodal human state and intention recognition, explainable artificial intelligence (XAI) techniques, and adaptive and learning-based control architectures. The analysis emphasizes the role of transparency, interpretability, and context-aware adaptation in establishing trust calibration within safety-critical and dynamic collaborative environments. A cross-sectional synthesis of the literature reveals several critical gaps, including the lack of standardized trust evaluation metrics, limited integration of explainability and control adaptation, inadequate consideration of long-term trust dynamics, and insufficient validation in real-world, unstructured environments. The analysis closes by delineating potential research avenues for cohesive, human-centered HRC frameworks that effortlessly incorporate trust modeling, explainable decision-making, and adaptive control. The ideas offered seek to inform the creation of advanced collaborative robots that are safe, efficient, transparent, adaptive, and trustworthy.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11554-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147738254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Thermohydraulic performance of spray cooling systems: a general model by machine learning","authors":"Mohammad Shamsodini Lori,&nbsp;Wenge Huang,&nbsp;Zhenhua Tian,&nbsp;Jiangtao Cheng","doi":"10.1007/s10462-026-11499-5","DOIUrl":"10.1007/s10462-026-11499-5","url":null,"abstract":"","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11499-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147561210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A review on vision-centric coarse to fine-grained animal action recognition","authors":"Ali Zia,&nbsp;Renuka Sharma,&nbsp;Abdelwahed Khamis,&nbsp;Usman Ali,&nbsp;Xuesong Li,&nbsp;Muhammad Husnain,&nbsp;Numan Shafi,&nbsp;Saeed Anwar,&nbsp;Imran Raza,&nbsp;Muhammad Hasan Jamal,&nbsp;Sabine Schmoelzl,&nbsp;Eric Stone,&nbsp;Lars Petersson,&nbsp;Vivien Rolland","doi":"10.1007/s10462-026-11526-5","DOIUrl":"10.1007/s10462-026-11526-5","url":null,"abstract":"<div><p>This review provides an in-depth exploration of the field of animal action recognition, focusing on coarse-grained (CG) and fine-grained (FG) techniques. The primary aim is to examine the current state of research in animal behaviour recognition and to elucidate the unique challenges associated with recognising subtle animal actions in outdoor environments. These challenges differ significantly from those encountered in human action recognition due to factors such as non-rigid body structures, frequent occlusions, and the lack of large-scale, annotated datasets. This review underscores the critical differences between human and animal action recognition. While inspired by progress in the human domain, animal action recognition presents unique challenges due to high intra-species variability, complex environmental interactions, and unstructured datasets that human-centric models cannot fully address. Recent multimodal frameworks such as ARTEMIS and MSQNet exemplify state-of-the-art progress by integrating textual cues derived from video with visual and audio modalities. When considered alongside established spatio-temporal architectures like SlowFast, these developments signal a shift toward richer multimodal paradigms in behaviour analysis. By assessing the strengths and weaknesses of current methodologies and introducing a recently published dataset, the review outlines future directions for advancing fine-grained action recognition, aiming to improve accuracy and generalisability in behaviour analysis across species. This review extends beyond earlier reviews by offering the first systematic treatment of coarse-grained (CG) and fine-grained (FG) action recognition in animals.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11526-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147642911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning approaches for asthma exacerbation predictions: a systematic review","authors":"Giovanna Cilluffo,&nbsp;Michele Atzeni,&nbsp;Velia Malizia,&nbsp;Martina Vettoretti,&nbsp;Gianluca Sottile","doi":"10.1007/s10462-026-11536-3","DOIUrl":"10.1007/s10462-026-11536-3","url":null,"abstract":"<div><p>Asthma exacerbations are critical events that can lead to severe health complications, hospitalizations, and increased healthcare costs. Accurate prediction of these exacerbations is essential for timely intervention and improved patient outcomes. Traditional statistical models face challenges in handling the high-dimensional nature of clinical and environmental data. In this context, machine learning (ML) techniques offer promising alternatives for predicting asthma exacerbations by leveraging diverse data sources, that are often high-dimensional, including electronic health records, environmental factors, and patient-reported outcomes. This systematic review evaluates the application of ML-style models, including the use of logistic regression, decision trees, gradient boosting machines, support vector machines, and deep learning approaches such as long short-term memory networks, for the prediction of asthma exacerbations. Our findings indicate that from model performance assessment point of view ensemble learning methods, particularly random forests and boosting, consistently achieve higher accuracy than the traditional statistical models. Moreover, neural networks and deep learning models show potential in capturing complex temporal dependencies associated with exacerbation risk. From a clinical perspective, the literature shows that traditional models such as logistic regression remain highly valued for their interpretability and alignment with clinical reasoning, allowing clinicians to identify actionable and biologically plausible risk factors for exacerbations. At the same time, more advanced ML approaches add clinical value by capturing temporal dynamics, environmental influences, and patient subgroups, but their adoption in practice depends critically on transparency and clear explanation of the drivers of risk. However, challenges remain, including model interpretability, generalizability across different populations, and integration into clinical practice. Future research should focus on enhancing explainability, improving data harmonization, and optimizing hybrid ML frameworks to develop robust predictive models for asthma management. This review highlights the need for interdisciplinary collaboration to translate ML advancements into clinically relevant applications, ultimately improving asthma care and reducing exacerbation-related morbidity.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 4","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11536-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147441676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Systematic literature review of multi-objective hyper-heuristics: a human-in-the-loop large language model methodology","authors":"Reza Ghanbarzadeh,&nbsp;Iman Ahadi Akhlaghi,&nbsp;Mahsa Ghafarian Gholamhossein,&nbsp;Muhammad Najeeb Khan,&nbsp;Seyedali Mirjalili","doi":"10.1007/s10462-026-11531-8","DOIUrl":"10.1007/s10462-026-11531-8","url":null,"abstract":"<div>\u0000 \u0000 <p>Multi-objective hyper-heuristics (MOHHs) have emerged as a powerful paradigm in computational intelligence, which enables the dynamic selection or generation of low-level heuristics to solve complex optimisation problems involving multiple objectives. Despite growing academic interest and a wide range of applications, there has been limited comprehensive analysis of the field’s evolution, methodologies, and challenges. This study presents a systematic literature review of 236 peer-reviewed publications on MOHHs published between 2005 and 2025, supported by a human-in-the-loop process that utilises large language models (LLMs) to assist screening and analysis. The review categorises application domains, characterises heuristic management strategies, maps learning mechanisms, and identifies emerging research themes. The findings reveal a marked shift from heuristic selection to generation-based and hybrid approaches, an increasing integration of reinforcement learning, and growing attention to adaptive, user-centric, and explainable optimisation. Methodological trends are also discussed in relation to benchmark use, performance evaluation, and theoretical grounding. The paper concludes with a thematic roadmap that outlines multiple future research directions, including LLM-guided MOHHs, many-objective optimisation, and preference-aware systems. This comprehensive review provides a foundation for advancing MOHH research and supports its application in challenging real-world problems.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11531-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147643015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Beyond the black box: lessons in explainability from AI in mammography","authors":"Andrea Ciardiello,&nbsp;Anna D’Angelo,&nbsp;Luigi De Angelis,&nbsp;Stefano Giagu,&nbsp;Evis Sala,&nbsp;Guido Gigante","doi":"10.1007/s10462-026-11518-5","DOIUrl":"10.1007/s10462-026-11518-5","url":null,"abstract":"<div><p>With AI already in clinical use, mammography serves as a critical test-bed for the challenges and potential of medical AI. However, its progress is hampered by the ‘black box’ nature of current AI algorithms, limiting clinician trust and transparency. This review analyses the field of Explainable AI (XAI) as a solution, examining its motivations, methods, and metrics. We find the field is dominated by post-hoc saliency methods that provide plausible but not necessarily faithful explanations of AI decision-making. This focus has led to an evaluation gap, where localization accuracy is used as a proxy for explanatory quality without verifying the model’s true reasoning. Inherently interpretable models that could offer more faithful insights are rarely implemented, and a lack of human-centred studies further obscures the clinical utility of current XAI techniques. We argue that for AI in mammography to realize its full potential, the field must urgently shift focus from creating plausible explanations to developing and validating inherently interpretable systems that provide faithful, clinically meaningful insights.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11518-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147643014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Graph neural networks for anomaly detection: a systematic review of dynamic temporal approaches","authors":"Fernando Ares-Robledo,&nbsp;Helena Rifà-Pous,&nbsp;Robert Clarisó","doi":"10.1007/s10462-026-11532-7","DOIUrl":"10.1007/s10462-026-11532-7","url":null,"abstract":"<div>\u0000 \u0000 <p>Graph Neural Network (GNN) have recently gained significant attention for their ability to model evolving relationships in graph-structured data, offering new opportunities for anomaly detection in cybersecurity. This Systematic Literature Review (SLR) examines the current state of research on these models applied to cybersecurity-related anomaly detection tasks. We systematically analyze 79 studies to identify key trends, challenges, and opportunities in this emerging field. Our review highlights that while GNN offer unique advantages such as capturing spatiotemporal dependencies and modeling complex cyber-threat patterns, several barriers remain, including scalability issues, limited real-world datasets, and the lack of interpretability in most models. Common architectures such as Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), and hybrid Transformer-based models are identified, along with their applications in intrusion detection, botnet detection, and anomaly detection in industrial control systems. Several studies propose that GNN can overcome current limitations by integrating contrastive learning and adaptive models for real-time threat detection. This review emphasizes the need for scalable, explainable, and deployment-ready solutions to fully realize the potential of dynamic graph models in cybersecurity. Future research should focus on developing scalable and adaptive architectures, integrating improved interpretability mechanisms, and enhancing model robustness through cross-domain validation and real-time applications.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11532-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147643012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intelligent traffic signal control based on reinforcement learning: a survey","authors":"Hang Xiao,&nbsp;Huale Li,&nbsp;Zhaobin Wang,&nbsp;Zhen Yang,&nbsp;Shuhan Qi,&nbsp;Jiajia Zhang,&nbsp;DingZhong Cai,&nbsp;JiaQi Yin","doi":"10.1007/s10462-026-11530-9","DOIUrl":"10.1007/s10462-026-11530-9","url":null,"abstract":"<div><p>Rapid urbanization and the surge in vehicle ownership have exacerbated traffic congestion, posing substantial economic, environmental, and social challenges. Traditional traffic signal control methods often struggle to address the dynamic complexities of modern urban traffic, frequently resulting in operational inefficiencies. Reinforcement Learning (RL), with its inherent capacity for real-time learning and adaptation, has emerged as a promising paradigm for optimizing Traffic Signal Control (TSC). RL approaches are particularly well-suited for handling complex traffic states and coordinating global optimization across multiple intersections. Despite notable progress, RL-based systems continue to face significant hurdles, including high computational costs, extensive data requirements, and issues regarding generalizability across diverse traffic scenarios. This paper synthesizes current RL-based models for TSC and highlights recent advancements in the field. It provides a comprehensive review of prominent approaches, categorizes existing studies based on their methodological frameworks, and conducts a technical evaluation of classical RL-based methods to assess their performance across varied traffic conditions. Finally, the remaining challenges and potential future directions for RL-based TSC are critically examined.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11530-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147606752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A survey on multimodal learning for industrial diagnostics: a data dimensionality perspective","authors":"Yangfeng Wang,&nbsp;Wenyong Yu,&nbsp;Changyang Yu,&nbsp;Hui Shi,&nbsp;Wenlong Li","doi":"10.1007/s10462-026-11527-4","DOIUrl":"10.1007/s10462-026-11527-4","url":null,"abstract":"<div>\u0000 \u0000 <p>Industrial diagnostics, as an essential methodology for quality control, production optimization, and operational safety assurance, has garnered global research efforts to enhance diagnostic efficacy. With advancements in hardware and deep learning technologies, multimodal learning has emerged as a transformative approach in this domain. Within multimodal frameworks, data inherently assumes a pivotal role, where inter-modal characteristics fundamentally determine diagnostic strategies. This study adopts an innovative data dimensionality perspective to investigate multimodal learning applications in industrial diagnostic tasks systematically. Through comprehensive surveys of cutting-edge research across task requirements, data characteristics, and fusion methodologies, we delineate the utilization patterns of heterogeneous multimodal data in diverse diagnostic scenarios. Furthermore, we systematically categorize feature extraction and fusion strategies based on the types of data modality and their intrinsic properties. The study ultimately identifies three core challenges and proposes actionable research directions to address these challenges. By elucidating technological advancements through data characteristics, this work provides a holistic understanding of state-of-the-art developments and practical guidelines for applied researchers seeking to implement multimodal solutions in real-world industrial settings.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 5","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11527-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147558762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}