arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

Artificial Intelligence Journal · 期刊 · Artificial Intelligence

共收录 152 篇
2609.29108 2026-09-25 cs.AI cs.SY eess.SY q-fin.GN q-fin.TR 新提交

Functional Architecture of European Electricity Trading Markets: Requirements for AI Supported Trading Systems under Regulatory Constraints

欧洲电力交易市场的功能架构:监管约束下AI支持交易系统的需求

Walter Kurz, Wojtek Stricker

机构 * Swissi Institute for AI(Swissi人工智能研究所)

AI总结 本文提出一种受监管约束的欧洲电力交易AI支持系统功能架构,通过决策状态向量、风险核算、约束优化和故障关闭控制,将AI作为有界决策组件嵌入市场运营。

Comments 12 pages, 1 table. Published in Swissi AI Journal under CC BY 4.0

Journal ref Swissi AI Journal, Volume 2026, Article SAIJ-cwo7xrcdsaut (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2609.27636 2026-09-24 q-fin.GN cs.MA q-fin.RM 新提交

Multi-Agent AI Architecture for Regulated Insurers: A generic AI framework under Solvency II and the AI Act in Austria and Germany

受监管保险公司的多智能体AI架构:奥地利和德国在偿付能力监管II与《人工智能法案》下的通用AI框架

Walter Kurz

AI总结 本文提出一种结合经济理论与制度设计的正式多智能体架构,用于在奥地利和德国受监管保险公司中实现合规、可审计的企业级AI,涵盖偿付能力监管II和《人工智能法案》等框架。

Comments 15 pages, 0 figures. Published in Swissi AI Journal under CC BY 4.0

Journal ref Swissi AI Journal, Volume 2025, Article SAIJ-qzvrl4bwy7y2 (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2609.27632 2026-09-24 q-fin.GN cs.AI 新提交

Compliant AI Infrastructure for Regulated Finance: A tiered multi-agent framework with DLT audit trails for financial operations in DACH

受监管金融的合规AI基础设施:面向DACH地区金融运营的具有DLT审计追踪的分层多智能体框架

Walter Kurz, Reinhard Magg

机构 * Swissi Institute for AI(瑞士人工智能研究所)

AI总结 提出面向受监管金融的合规优先AI架构,将监管作为定位层,通过策略编译器映射禁令与义务,结合许可DAG审计追踪和条款级索引,实现可验证的嵌入式合规。

Comments 16 pages, 3 figures. Published in Swissi AI Journal under CC BY 4.0

Journal ref Swissi AI Journal, Volume 2025, Article SAIJ-xz3bi3q7fwim (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2609.24181 2026-09-22 q-fin.GN econ.GN q-fin.EC 新提交

Firm Valuation When AI Shapes the Business Model: A Milestone-Based Real-Options Framework for the AI Valuation Uncertainty Problem

AI塑造商业模式时的企业估值:针对AI估值不确定性的基于里程碑的实物期权框架

Walter Kurz, Wojtek Stricker, Stefan Marx, Frank Reinhardt, Florian Kollberg

AI总结 针对AI估值不确定性,提出基于里程碑的实物期权框架,区分AI整合者与提供者,分解期权假设,应用于AI能源SaaS公司产生可追溯估值区间,风险集中于后期延续期权。

Comments 19 pages, 0 figures. Published in Swissi AI Journal under CC BY 4.0

Journal ref Swissi AI Journal, Volume 2026, Article SAIJ-f3jtignfyge3 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.28024 2026-06-29 cs.AI cs.LG 新提交

Lifted Causal Inference

提升因果推断

Malte Luttermann, Tanya Braun, Ralf Möller, Marcel Gehrke

机构 * German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心)

AI总结 提出参数化因果因子图(PCFG)和提升因果推断(LCI)算法,在关系域中高效计算因果效应,显著加速推理。

Comments Accepted to the Annals of Mathematics and Artificial Intelligence journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.22354 2026-06-23 cs.RO 版本更新

LLM-Based Generalizable Hierarchical Task Planning and Execution for Heterogeneous Robot Teams with Event-Driven Replanning

基于LLM的异构机器人团队通用分层任务规划与执行及事件驱动重规划

Suraj Borate, Bhavish Rai B, Vipul Pardeshi, Madhu Vadali

机构 * Department of Mechanical Engineering, IIT Gandhinagar(印度加尔各直辖区理工学院机械工程系) ; Sahyadri College of Engineering and Management(萨哈亚德里工程与管理学院) ; Vishwakarma Institute of Information Technology(维什瓦克arma信息技术学院)

AI总结 提出CoMuRoS架构,结合集中式规划与分布式执行,通过LLM解释自然语言目标、分配子任务,并支持事件驱动重规划,在硬件实验中实现自主恢复和协调运输。

Comments full version of this short paper is accepted at Frontiers in Robotics and AI Journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.07530 2026-06-09 cs.CL 新提交

Finding New Connections between Concepts from Medline Database Incorporating Domain Knowledge

从Medline数据库中结合领域知识发现概念间的新连接

Yang Weikang, Chowdhury S. M. Mazharul Hoque, Jin Wei

AI总结 提出一种基于Swanson ABC模型的改进自适应模型,用于文献发现中隐藏的概念连接,通过中间主题B连接看似无关的主题A和C。

Journal ref Artificial Intelligence, IntechOpen, 2024

详情

展开后加载摘要…

URL PDF HTML 收藏
2408.04207 2026-06-09 cs.NE 版本更新

Proven Advantage of Multiobjective Evolutionary Algorithms for Problems with Different Degrees of Conflict

多目标进化算法在冲突程度不同问题中的优势证明

Weijie Zheng

AI总结 本文通过对比不同冲突程度的问题,证明多目标进化算法在解决OneMaxMin_k问题时优于传统方法,且在不同冲突程度下均能高效求解。

Comments This is the preprint version of an article accepted for publication in Artificial Intelligence Journal

Journal ref Artificial Intelligence, 2026, 104573

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.24543 2026-05-26 cs.AI cs.SY eess.SY

Emission-Aware Reinforcement Learning for Sustainable Electric Vehicle Charging and Carbon Dioxide Reduction Under Varying Renewable Penetration

面向可持续电动汽车充电与二氧化碳减排的排放感知强化学习:在不同可再生能源渗透率下

Ninglin Ou, Mohammad A. Razzaque, Iftekher Islam Shovon, Shafkat Khan Siam, Shafiuzzaman K Khadem, Krishnendu Guha, Mayeen U Khandaker, Md. Noor-A-Rahim

机构 * organization= nasc Research, School of Computer Science \& IT, University College Cork , country = IE ; organization= School of Computing, Engineering ; Digital Technologies, Teesside University , country= UK ; organization= International Energy Research Centre, Tyndall National Institute, Cork , country = IE ; Radiation Technologies Group, CCDCU, Faculty of Engineering ; Technology, Sunway University , country = Malaysia ; organization= Department of Physics, College of Science, Korea University , country = Republic of Korea

AI总结 提出基于软演员-评论家算法的排放感知强化学习策略,通过多目标奖励函数优化电动汽车充电调度,在EV2Gym平台上实现高达87%的碳排放减少和52%的可再生能源自消纳率。

Comments Submitted the Engineering Applications of Artificial Intelligence Journal (Elsevier)

详情

展开后加载摘要…

URL PDF HTML 收藏
2105.13431 2026-05-26 cs.LG cs.AI cs.SY eess.SY

An Offline Risk-aware Policy Selection Method for Bayesian Markov Decision Processes

贝叶斯马尔可夫决策过程的离线风险感知策略选择方法

Giorgio Angelotti, Nicolas Drougard, Caroline Ponzoni Carvalho Chanel

机构 * Natural Intelligence Toulouse Institute, University of Toulouse, France(图卢兹大学自然智能研究所) ; ISAE-SUPAERO, University of Toulouse, France(图卢兹大学ISAE-SUPAERO)

AI总结 针对离线强化学习中模型不确定性导致策略风险高的问题,提出一种基于贝叶斯形式化框架的风险感知策略选择方法EvC,通过最大化贝叶斯后验下的风险感知目标来选择稳健策略。

Comments Preprint, under review

Journal ref Artificial Intelligence, Volume 354, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
cs/0507027 2026-03-05 cs.GT cs.CC cs.MA

Anyone but Him: The Complexity of Precluding an Alternative

任何人但 him:排除替代品的复杂性

Edith Hemaspaandra, Lane A. Hemaspaandra, Joerg Rothe

AI总结 本文研究了在多智能体环境下,通过机制如选民/候选人添加/删除/分割等,防止特定候选人获胜的复杂性问题,发现不同选举系统对破坏性控制的防护能力各不相同。

Comments This revision--the March 2026 Version 5--is identical to the March 2006 Version 4 except in providing, as Appendix A, a correction to the second half of the proof of Theorem 4.21 as it appears in both Version 4 and the AIJ journal version; this proof also replaces the analogous proof part of Theorem 6 of the AAAI version

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.10584 2026-02-12 cs.LG

When Gradient Clipping Becomes a Control Mechanism for Differential Privacy in Deep Learning

当梯度裁剪成为深度学习中差分隐私的控制机制

Mohammad Partohaghighi, Roummel Marcia, Bruce J. West, YangQuan Chen

机构 * Computer Science, University of California, Merced, CA 95343, USA(计算机科学,加州大学默塞德分校) ; Department of Applied Mathematics, University of California, Merced, CA 95343, USA(应用数学系,加州大学默塞德分校) ; Department of Innovation(创新部门) ; Research, North Carolina State University, Raleigh, NC, USA(研究,北卡罗来纳州立大学,拉洛城) ; Automation (MESA) Lab, Department of Mechanical Engineering, School of Engineering, University of California, Merced, CA 95343, USA(自动化(MESA)实验室,机械工程系,工程学院,加州大学默塞德分校)

AI总结 本文提出了一种基于控制机制的梯度裁剪策略,通过模型参数的谱诊断动态调整裁剪阈值,以在差分隐私训练中平衡隐私保护与模型性能。

Comments This manuscript is under review in the Engineering Applications of Artificial Intelligence journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.09798 2026-02-11 cs.AI

Symbolic Pattern Temporal Numeric Planning with Intermediate Conditions and Effects

具有中间条件和效果的符号模式时间数字规划

Matteo Cardellini, Enrico Giunchiglia

AI总结 本文提出了一种扩展的符号模式规划方法,用于处理具有中间条件和效果的时间规划问题,实验显示其在多个领域表现优异。

Comments Under review at the Artificial Intelligence Journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2402.10192 2026-02-05 cs.LG cs.AI cs.DM quant-ph

Multi-Excitation Projective Simulation with a Many-Body Physics Inspired Inductive Bias

多激发投影模拟:受多体物理启发的归纳偏置

Philip A. LeMaitre, Marius Krumm, Hans J. Briegel

机构 * University of Innsbruck, Institute for Theoretical Physics(因斯布鲁克大学理论物理研究所)

AI总结 本文提出多激发投影模拟(mePS)方法,通过受多体物理启发的归纳偏置解决超图建模中的指数复杂性问题,并在多个场景中验证了其资源节省和可解释性优势。

Comments 41 pages, 9 figures; Code repository at https://github.com/MariusKrumm/ManyBodyMEPS. Updated to be consistent with AIJ version

Journal ref Artificial Intelligence 352, 104489 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.22183 2026-02-02 cs.DB cs.AI cs.DS

COL-Trees: Efficient Hierarchical Object Search in Road Networks

COL-Trees: 路网中高效的层次对象搜索

Tenindra Abeywickrama, Muhammad Aamir Cheema, Sabine Storandt

机构 * Center for Computational Science, RIKEN(计算科学中心,RIKEN) ; Faculty of Information Technology, Monash University(信息科技学院,墨尔本大学) ; Department of Computer and Information Science, University of Konstanz(计算机与信息科学系,康斯坦茨大学)

AI总结 COL-Trees通过高效的层次图遍历方法,提升了道路网络中多代理POI搜索的效率,实现高达4个数量级的性能改进。

Comments Submitted to Artificial Intelligence (AIJ)

详情

展开后加载摘要…

URL PDF HTML 收藏
2307.15586 2026-01-27 cs.GT econ.TH

Settling the Score: Portioning with Cardinal Preferences

解决得分问题:基于基数偏好的分割

Edith Elkind, Matthias Greger, Patrick Lederer, Warut Suksompong, Nicholas Teh

AI总结 本文研究了基于基数偏好的分割问题,提出通过计算提案平均值的简单规则,在公平性方面表现优异,并从公理学角度进行了详细分析。

Comments A preliminary version appeared in the 26th European Conference on Artificial Intelligence (ECAI), 2023

Journal ref Artificial Intelligence, 352:104487 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.12286 2026-01-21 cs.LO cs.AI

SHACL Validation in the Presence of Ontologies: Semantics and Rewriting Techniques

在本体存在的情况下SHACL验证的语义与重写技术

Anouk Oudshoorn, Magdalena Ortiz, Mantas Simkus

AI总结 本文提出了一种基于核心普遍模型的SHACL验证语义,利用描述逻辑Horn-ALCHIQ构建本体模型,并通过重写技术将本体存在下的SHACL验证转化为标准验证,证明其复杂性为EXPTIME完全问题。

Comments Published in AIJ

Journal ref Volume 352, 2026, 104483

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.05637 2026-01-14 cs.LG cs.AI cs.CR

Federated Neural Nonparametric Point Processes

联邦神经非参数点过程

Hui Chen, Xuhui Fan, Hengyu Liu, Yaqiong Li, Zhilin Zhao, Feng Zhou, Christopher John Quinn, Longbing Cao

机构 * Macquarie University(麦考瑞大学) ; Aalborg University(奥胡斯大学) ; Renmin University of China(中国人民大学) ; Iowa State University(爱荷华州立大学)

AI总结 FedPP通过整合神经嵌入和SGCPs,有效解决联邦系统中稀疏和不确定事件建模问题,提升隐私保护与个性化性能。

Journal ref Artificial Intelligence, vol. 351, 104454, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.15261 2025-11-18 cs.GT cs.DS math.PR

Pandora's Box Problem With Time Constraints

带时间约束的潘多拉魔盒问题

Georgios Amanatidis, Ben Berger, Tomer Ezra, Michal Feldman, Federico Fusco, Rebecca Reiffenhäuser, Artem Tsikiridis

AI总结 针对评估成本高昂时寻找最佳替代方案的潘多拉魔盒问题,本文引入涵盖时间作用变体的通用框架,提供NP难问题的常数因子近似策略,并针对无处理时间等特殊情况获改进结果。

Comments This paper unifies and extends preliminary versions that appeared in AAAI 2024 (DOI:10.1609/aaai.v38i18.30015) and WINE 2024 (arXiv:2407.15261v1)

Journal ref Artificial Intelligence, Vol. 349, 104426, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.03417 2025-10-01 cs.LG cs.AI

Learning Semantic Association Rules from Internet of Things Data

从物联网数据中学习语义关联规则

Erkan Karabulut, Paul Groth, Victoria Degeler

机构 * University of Amsterdam(阿姆斯特丹大学)

AI总结 本文针对物联网数据的异构性与大规模问题,提出融合动态传感器数据和静态元数据的ARM流程,并用基于自动编码器的Aerial提取更简洁且全覆盖的高质量关联规则。

Comments This paper has been published in Neurosymbolic Artificial Intelligence Journal (SAGE)

详情

展开后加载摘要…

URL PDF HTML 收藏
2302.06615 2025-09-10 cs.AI cs.LG

Self-Emotion-Mediated Exploration in Artificial Intelligence Mirrors: Findings from Cognitive Psychology

人工智能镜像中的自我情绪介导探索:来自认知心理学的发现

Gustavo Assunção, Miguel Castelo-Branco, Paulo Menezes

机构 * University of Coimbra(科英布拉大学) ; Institute for Biomedical Imaging and Translational Research(生物医学影像与转化研究 institutes) ; Institute of Systems and Robotics(系统与机器人研究所)

AI总结 本文提出一种基于认知与成就情绪的双模块强化学习框架,赋予AI代理内在探索驱动力,实验显示惊讶与探索正相关、自豪负相关,验证了生物启发对AI自主性的价值。

Comments 17 pages, 8 figures, MDPI's AI Journal

Journal ref Assuncao, G.; Castelo-Branco, M.; Menezes, P. Self-Emotion-Mediated Exploration in Artificial Intelligence Mirrors: Findings from Cognitive Psychology. AI 2025, 6, 220

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.19372 2025-07-28 cs.AI cs.LG

Learning neuro-symbolic convergent term rewriting systems

学习神经符号收敛项重写系统

Flavio Petruzzellis, Alberto Testolin, Alessandro Sperduti

机构 * Department One(部门一) ; Department Two(部门二) ; Department of Mathematics, University of Padova(帕多瓦大学数学系) ; Department of General Psychology, University of Padova(帕多瓦大学普通心理学系) ; Augmented Intelligence Center, Fondazione Bruno Kessler(布鲁诺·凯塞勒基金会增强智能中心)

AI总结 本文提出受重写算法启发的神经符号框架及 NRS、FastNRS,用于学习收敛项重写系统,并在公式化简和多领域任务中展现强分布外泛化,性能优于强神经基线并追平或超过 o1-preview。

Comments 48 pages, 31 figures. Submitted for review by Artificial Intelligence Journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.01558 2025-07-21 cs.HC cs.AI

Visual Grounding Methods for Efficient Interaction with Desktop Graphical User Interfaces

面向桌面图形用户界面高效交互的视觉定位方法

El Hassane Ettifouri, Jessica López Espejel, Laura Minkova, Tassnim Dardouri, Walid Dahhane

机构 * Research and Innovation Lab, Novelis(创新研究实验室,Novelis)

AI总结 本文针对GUI合成图像视觉定位不足的问题,提出IVGocr和IVGdirect两种指令视觉定位方法,并构建专用数据集及CPV评估指标,以提升AI智能体与桌面界面的交互效率。

Comments Preprint submitted to Engineering Applications of Artificial Intelligence journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2006.00025 2025-06-18 cs.RO

Environmental regulation using Plasticoding for the evolution of robots

使用Plasticoding的环境调控用于机器人进化

Karine Miras, Eliseo Ferrante, A. E. Eiben

AI总结 针对进化机器人系统仅通过选择间接受环境影响的问题,提出基于Plasticoding的编码方法,让机器人可依环境刺激在生命周期内调整形态、控制器与行为,实验验证其能提升任务性能并产生多样进化结果。

Comments This paper was submitted to the Frontiers in Robotics and AI journal on the 22/02/2020, and is still under review

详情

展开后加载摘要…

URL PDF HTML 收藏
2211.12738 2025-06-17 cs.GT cs.CR

Differentially Private Fair Division

差分隐私公平分配

Pasin Manurangsi, Warut Suksompong

AI总结 该研究聚焦不可分割资源公平分配的隐私问题,基于差分隐私框架提出满足近似无嫉妒性与比例性的算法,同时给出不同相邻实例定义下两类公平准则的强否定结果。

Comments Appears in the 37th AAAI Conference on Artificial Intelligence (AAAI), 2023

Journal ref Artificial Intelligence, 347:104385 (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.01156 2025-05-22 cs.LG

Machine Learning for Physical Simulation Challenge Results and Retrospective Analysis: Power Grid Use Case

物理模拟机器学习挑战赛结果与回顾性分析:电网用例

Milad Leyli-Abadi, Jérôme Picault, Antoine Marot, Jean-Patrick Brunet, Agathe Gilain, Amarsagar Reddy Ramapuram Matavalam, Shaban Ghias Satti, Qingbin Jiang, Yang Liu, Dean Justin Ninalga

AI总结 本文介绍ML4PhySim竞赛,利用LIPS框架评估AI加速电网潮流模拟方法,在30%可再生能源模型上实现至少一个数量级加速,并保持可靠性,优于传统模拟。

Comments 47 pages, 12 figures, 4 table, submission to Energy and AI journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2404.07678 2025-05-06 cs.CY econ.GN q-fin.EC

On the role of ethics and sustainability in business innovation

论伦理与可持续性在商业创新中的作用

Maria Fay, Frederik F. Flöther

AI总结 本文探讨了伦理与可持续性在商业创新中的作用,分析了现有创新采用框架的不足,提出通过系统思维等方法促进组织创新文化的发展,特别关注新兴技术如人工智能的伦理和可持续性考量。

Journal ref Artificial Intelligence, Entrepreneurship and Risk. Springer, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2308.11738 2025-02-27 cs.AI cs.LO math.CO

Lifted Inference beyond First-Order Logic

超越一阶逻辑的提升推理

Sagar Malhotra, Davide Bizzaro, Luciano Serafini

机构 * University of Padova(帕多瓦大学)

AI总结 本研究扩展了C^2逻辑片段的域可提升性,使其能处理有向无环图、连通图、树和森林等现实数据属性,并提出“通过分裂计数”的新方法论,为概率推理和组合结构计数提供通用框架。

Comments Explanation for practical implementation of cardinality constraints added in Appendix .arXiv admin note: text overlap with arXiv:2302.09830

Journal ref Artificial Intelligence,Volume 342,2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2306.02611 2025-02-18 cs.NE cs.AI

Stochastic Population Update Can Provably Be Helpful in Multi-Objective Evolutionary Algorithms

随机种群更新可被证明对多目标进化算法有益

Chao Bian, Yawen Zhou, Miqing Li, Chao Qian

机构 * Nanjing University(南京大学) ; University of Birmingham(伯明翰大学)

AI总结 针对多目标进化算法(MOEAs)种群更新通常为贪婪确定性的问题,本文证明将其替换为随机机制可使SMS-EMOA和NSGA-II求解两个双目标问题的期望运行时间指数级减少,实证亦验证该方法有效性,为MOEAs开发提供新方向。

Comments Accepted by Artificial Intelligence Journal. A preliminary version of this paper has appeared at IJCAI'23

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.06123 2025-02-17 cs.LG

Contrastive Federated Learning with Tabular Data Silos

面向表格数据孤岛的对比联邦学习

Achmad Ginanjar, Xue Li, Wen Hua, Jiaming Pei

机构 * The University of Queensland(昆士兰大学) ; Monash University(莫纳什大学) ; Indonesia Taxation Office(印度尼西亚税务局) ; Queensland University of Technology(昆士兰科技大学) ; SUNY Albany(纽约州立大学奥尔巴尼分校) ; Dartmouth College(达特茅斯学院) ; Hong Kong Polytechnic University(香港理工大学) ; Renmin University of China(中国人民大学)

AI总结 针对垂直划分数据孤岛的样本错位与隐私问题,提出对比联邦学习(CFL),通过本地获取对比表示并联邦聚合知识,在不共享数据的前提下提升模型性能,优于现有方法。

Comments 44 Pages. 1stversion was submitted on Artificial Intelligence Journal, Jan 29, 2024, ARTINT-D-24-00098

详情

展开后加载摘要…

URL PDF HTML 收藏
↑