检索结果-内蒙古大学图书馆

Fairness in Algorithmic Profiling: The AMAS Case

MINDS AND machineS 2025年第1期35卷 1-30页

作者： Achterhold, Eva Muehlboeck, Monika Steiber, Nadia Kern, Christoph Ludwig Maximilians Univ Munchen Munich Germany Univ Vienna Vienna Austria Munich Ctr Machine Learning MCML Munich Germany

We study a controversial application of algorithmic profiling in the public sector, the Austrian AMAS system. AMAS was supposed to help caseworkers at the Public Employment Service (PES) Austria to allocate support measures to job seekers based on their predicted chance of (re-)integration into the labor market. Shortly after its release, AMAS was criticized for its apparent unequal treatment of job seekers based on gender and citizenship. We systematically investigate the AMAS model using a novel real-world dataset of young job seekers from Vienna, which allows us to provide the first empirical evaluation of the AMAS model with a focus on fairness measures. We further apply bias mitigation strategies to study their effectiveness in our real-world setting. Our findings indicate that the prediction performance of the AMAS model is insufficient for use in practice, as more than 30% of job seekers would be misclassified in our use case. Further, our results confirm that the original model is biased with respect to gender as it tends to (incorrectly) assign women to the group with high chances of re-employment, which is not prioritized in the PES' allocation of support measures. However, most bias mitigation strategies were able to improve fairness without compromising performance and thus may form an important building block in revising profiling schemes in the present context.

关键词： Algorithmic profiling Statistical discrimination Public employment services Artificial intelligence Bias mitigation

来源：评论

学校读者我要写书评

暂无评论

MINIMAX PROBLEMS FOR ENSEMBLES OF CONTROL-AFFINE SYSTEMS

引用

SIAM JOURNAL ON CONTROL AND OPTIMIZATION 2025年第1期63卷 502-523页

作者： Scagliotti, Alessandro Tech Univ Munich CIT Sch Boltzmannstr 3-II D-85748 Garching Germany Munich Ctr Machine Learning MCML Munich Germany

In this paper, we consider ensembles of control-affine systems in IIBd, and we study simultaneous optimal control problems related to the worst-case minimization. After proving that such problems admit solutions, denoting with (\ThetaN)N a sequence of compact sets that parametrize the ensembles of systems, we first show that the corresponding minimax optimal control problems are \Gamma-convergent whenever (\ThetaN)N has a limit with respect to the Hausdorff distance. Besides its independent interest, the previous result plays a crucial role for establishing the Pontryagin Maximum Principle (PMP) when the ensemble is parametrized by a set \Theta consisting of infinitely many points. Namely, we first approximate \Theta by finite and increasing-in-size sets (\ThetaN)N for which the PMP is known, and then we derive the PMP for the \Gamma-limiting problem. The same strategy can be pursued in applications, where we can reduce infinite ensembles to finite ones to compute the minimizers numerically. We bring as a numerical example the Schro"\dinger equation for a qubit with uncertain resonance frequency.

关键词： simultaneous control minimax optimal control Gamma-convergence Pontryagin Maxi- mum Principle

来源：评论

学校读者我要写书评

暂无评论

From NeurODEs to AutoencODEs: A mean-field control framework for width-varying neural networks

引用

EUROPEAN JOURNAL OF APPLIED MATHEMATICS 2025年第2期36卷 188-230页

作者： Cipriani, Cristina Fornasier, Massimo Scagliotti, Alessandro Tech Univ Munich Sch Computat Informat & Technol Munich Germany Munich Data Sci Inst MDSI Munich Germany Munich Ctr Machine Learning MCML Munich Germany

The connection between Residual Neural Networks (ResNets) and continuous-time control systems (known as NeurODEs) has led to a mathematical analysis of neural networks, which has provided interesting results of both theoretical and practical significance. However, by construction, NeurODEs have been limited to describing constant-width layers, making them unsuitable for modelling deep learning architectures with layers of variable width. In this paper, we propose a continuous-time Autoencoder, which we call AutoencODE, based on a modification of the controlled field that drives the dynamics. This adaptation enables the extension of the mean-field control framework originally devised for conventional NeurODEs. In this setting, we tackle the case of low Tikhonov regularisation, resulting in potentially non-convex cost landscapes. While the global results obtained for high Tikhonov regularisation may not hold globally, we show that many of them can be recovered in regions where the loss function is locally convex. Inspired by our theoretical findings, we develop a training method tailored to this specific type of Autoencoders with residual connections, and we validate our approach through numerical experiments conducted on various examples.

关键词： machine learning optimal control gradient flow minimising movement scheme autoencoders

来源：评论

学校读者我要写书评

暂无评论

An equivalency and efficiency study for one year digital pathology for clinical routine diagnostics in an accredited tertiary academic center

引用

VIRCHOWS ARCHIV 2025年 1-10页

作者： Iwuajoku, Viola Ekici, Kuebra Haas, Anette Khan, Mohammed Zaid Kazemi, Azar Kasajima, Atsuko Delbridge, Claire Muckenhuber, Alexander Schmoeckel, Elisa Stoegbauer, Fabian Bollwein, Christine Schwamborn, Kristina Steiger, Katja Mogler, Carolin Schueffler, Peter J. Tech Univ Munich Inst Pathol Munich Germany Munich Data Sci Inst MDSI Munich Germany Munich Ctr Machine Learning MCML Munich Germany

Digital pathology is revolutionizing clinical diagnostics by offering enhanced efficiency, accuracy, and accessibility of pathological examinations. This study explores the implementation and validation of digital pathology in a large tertiary academic center, focusing on its gradual integration and transition into routine clinical diagnostics. In a comprehensive validation process over a 6-month period, we compared sign-out of digital and physical glass slides of a wide range of different tissue specimens and histopathological diagnoses. Key metrics such as diagnostic concordance and user satisfaction were assessed by involving the pathologists in a validation training and study phase. We measured turnaround times before and after transitioning to digital pathology to assess the impact on overall efficiency. Our results demonstrate a 99% concordance between the analog and digital reports while at the same time reducing the time to sign out a case by almost a minute, suggesting potential long-term efficiency gains. Our digital transition positively impacted our pathology workflow: Pathologists reported increased flexibility and satisfaction due to the ease of accessing and sharing digital slides. However, challenges were identified, including technical issues related to image quality and system integration. Lessons learned from this study emphasize the importance of robust training programs, adequate IT support, and ongoing evaluation to ensure successful integration. This validation study confirms that digital pathology is a viable and beneficial tool for accurate clinical routine diagnostics in large academic centers, offering insights for other institutions considering similar endeavors.

关键词： Digital pathology Digital transition Validation study Efficiency

来源：评论

学校读者我要写书评

暂无评论

Normalizing flows as approximations of optimal transport maps via linear-control neural ODEs

引用

NONLINEAR ANALYSIS-THEORY METHODS & APPLICATIONS 2025年 257卷

作者： Scagliotti, A. Farinelli, S. Tech Univ Munich CIT Sch Boltzmannstr 3-II D-85748 Garching Germany Munich Ctr Machine Learning MCML Munich Germany Univ Genoa DIMA MalGa Via Dodecaneso 35 I-16146 Genoa Italy

In this paper, we consider the problem of recovering the W-2-optimal transport map T between absolutely continuous measures mu, nu is an element of P(R-n) as the flow of a linear-control neural ODE, where the control depends only on the time variable and takes values in a finite-dimensional space. We first show that, under suitable assumptions on mu, nu and on the controlled vector fields governing the neural ODE, the optimal transport map is contained in the C-c(0)-closure of the flows generated by the system. Then, we tackle the problem under the assumption that only discrete approximations of mu(N), nu(N) of the original measures mu, nu are available: we formulate approximated optimal control problems, and we show that their solutions give flows that approximate the original optimal transport map T. In the framework of generative models, the approximating flow constructed here can be seen as a 'Normalizing Flow', which usually refers to the task of providing invertible transport maps between probability measures by means of deep neural networks. We propose an iterative numerical scheme based on the Pontryagin Maximum Principle for the resolution of the optimal control problem, resulting in a method for the practical computation of the approximated optimal transport map, and we test it on a two-dimensional example.

关键词： Optimal transport Optimal control Gamma-convergence Linear-control neural ODEs

来源：评论

学校读者我要写书评

暂无评论

CromSS: Cross-Modal Pretraining With Noisy Labels for Remote Sensing Image Segmentation

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Liu, Chenying Albrecht, Conrad M. Wang, Yi Zhu, Xiao Xiang Tech Univ Munich TUM Chair Data Sci Earth Observat D-80333 Munich Germany German Aerosp Ctr DLR Remote Sensing Technol Inst D-82234 Wessling Germany Munich Ctr Machine Learning MCML D-80538 Munich Germany Remote Sensing Technol Inst DLR D-82234 Wessling Germany TUM Chair Data Sci Earth Observat D-80333 Munich Germany MCML D-80538 Munich Germany

We explore the potential of large-scale noisily labeled data to enhance feature learning by pretraining semantic segmentation models within a multimodal framework for geospatial applications. We propose a novel cross-modal sample selection (CromSS) method, a weakly supervised pretraining strategy designed to improve feature representations through cross-modal consistency and noise mitigation techniques. Unlike conventional pretraining approaches, CromSS exploits massive amounts of noisy and easy-to-come-by labels for improved feature learning beneficial to semantic segmentation tasks. We investigate middle and late fusion strategies to optimize the multimodal pretraining architecture design. We also introduce a cross-modal sample selection module to mitigate the adverse effects of label noise, which employs a cross-modal entangling strategy to refine the estimated confidence masks within each modality to guide the sampling process. Additionally, we introduce a spatial-temporal label smoothing technique to counteract overconfidence for enhanced robustness against noisy labels. To validate our approach, we assembled the multimodal dataset, NoLDO-S12, which consists of a large-scale noisy label subset from Google's Dynamic World (DW) dataset for pretraining and two downstream subsets with high-quality labels from Google DW and OpenStreetMap (OSM) for transfer learning. Experimental results on two downstream tasks and the publicly available DFC2020 dataset demonstrate that when effectively utilized, the low-cost noisy labels can significantly enhance feature learning for segmentation tasks. The data, codes, and pretrained weights are freely available at https://***/zhu-xlab/CromSS.

关键词： Noise measurement Noise Representation learning Remote sensing Training Deep learning Semantic segmentation Transfer learning Earth Smoothing methods Geospatial artificial intelligence multimodal deep learning noisy labels pretraining sample selection semantic segmentation SSL4EO-S12 dataset

来源：评论

学校读者我要写书评

暂无评论

Toward Integrating ChatGPT Into Satellite Image Annotation Workflows: A Comparison of Label Quality and Costs of Human and Automated Annotators

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2025年 18卷 4366-4381页

作者： Beck, Jacob Kemeter, Lukas Malte Duerrbeck, Konrad Abdalla, Mohamed Hesham Ibrahim Kreuter, Frauke Ludwig Maximilians Univ Munchen Munich Ctr Machine Learning MCML Inst Informat D-80538 Munich Germany Fraunhofer Inst Integrated Circuits IIS Ctr Appl Res Supply Chain Serv D-90411 Nurnberg Germany

High-quality annotations are a critical success factor for machine learning (ML) applications. To achieve this, we have traditionally relied on human annotators, navigating the challenges of limited budgets and the varying task-specific expertise, costs, and availability. Since the emergence of large language models (LLMs), their popularity for generating automated annotations has grown, extending possibilities and complexity of designing an efficient annotation strategy. Increasingly, computer vision capabilities have been integrated into general-purpose LLMs like ChatGPT. This raises the question of how effectively LLMs can be used in satellite image annotation tasks and how they compare to traditional annotator types. This study presents a comprehensive investigation and comparison of various human and automated annotators for image classification. We evaluate the feasibility and economic competitiveness of using the ChatGPT4-V model for a complex land usage annotation task and compare it with alternative human annotators. A set of satellite images is annotated by a domain expert and 15 additional human and automated annotators, differing in expertise and costs. Our analyzes examine the annotation quality loss between the expert and other annotators. This comparison is conducted through, first, descriptive analyzes, second, fitting linear probability models, and third, comparing F1-scores. Ultimately, we simulate annotation strategies where samples are split according to an automatically assigned certainty score. Routing low-certainty images to human annotators can cut total annotation costs by over 50% with minimal impact on label quality. We discuss implications regarding the economic competitiveness of annotation strategies, prompt engineering, and the task-specificity of expertise.

关键词： Annotations Satellite images Biological system modeling Costs Chatbots Training Image annotation Visualization Remote sensing Data models Automated annotations ChatGPT label quality large language models (LLMs) satellite image annotation

来源：评论

学校读者我要写书评

暂无评论

Optimizing lower extremity CT angiography: A prospective study of individualized vs. fixed post-trigger delays in bolus tracking

引用

EUROPEAN JOURNAL OF RADIOLOGY 2025年 185卷 112009-112009页

作者： Nas, Loran Hoppe, Boj F. Stueber, Anna T. Grosu, Sergio Fink, Nicola von Fragstein, Alina Rudolph, Jan Ricke, Jens Sabel, Bastian O. Ludwig Maximilians Univ Munchen LMU Univ Hosp Dept Radiol Munich Germany Munich Ctr Machine Learning MCML Geschwister Scholl Pl 1 D-80539 Munich Germany Ludwig Maximilians Univ Munchen Dept Stat Ludwigstr 33 D-80539 Munich Germany

Purpose: To compare the contrast media opacification and diagnostic quality in lower-extremity runoff CT angiography (CTA) between bolus-tracking using conventional fixed trigger delay and patient-specific individualized post-trigger delay. Methods: In this prospective study, lower-extremity runoff CTA was performed in two cohorts, using either fixed or individualized trigger delay. Both cohorts had identical CT protocols, contrast media applications, and image reconstructions. Objective image quality (mean contrast opacification in HU), and subjective image quality (5point Likert-scale), were assessed in six vessels: abdominal aorta (AA), common iliac artery (CIA), superficial femoral artery (SFA), popliteal artery (PA), posterior tibial artery (PTA), and dorsalis pedis artery (DPA) by one rater for objective and two raters for subjective image quality. Objective image quality was analyzed using Student t-tests, while subjective ratings were compared with Fisher's exact test. Results: Overall, 65 patients were included (mean age: 71 f 14;39 men), 35 in the individualized cohort and 30 in the fixed cohort. No differences were found between the groups regarding demographics or radiation exposure. Individualized trigger delay ranged from 2 to 23 s (mean: 8.7 f 4.0 s) and was 10 s in the fixed cohort. The individualized cohort showed higher opacification in the peripheral arteries (PTA: 479 f 140 HU vs. 379 f 106 HU;p = 0.009;DPA: 477 f 191 HU vs. 346 f 137 HU;p = 0.009). Overall subjective "image quality" was rated higher in the individualized group ("excellent" or "good" in Rater 1: 97 % vs. 57 %;p <0.001;and Rater 2: 89 % vs. 53 %;p = 0.002). Conclusion: Individualized post-trigger delay enhances diagnostic quality, by improving vessel opacification in peripheral arteries and increasing subjective image quality in lower extremity runoff CTA.

关键词： Computed Tomography Angiography Peripheral Artery Diseases Lower Extremity Algorithms

来源：评论

学校读者我要写书评

暂无评论

VoxNeRF: Bridging Voxel Representation and Neural Radiance Fields for Enhanced Indoor View Synthesis

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2025年第6期10卷 5903-5910页

作者： Wang, Sen Cheng, Qing Gasperini, Stefano Zhang, Wei Wu, Shun-Cheng Zeller, Niclas Cremers, Daniel Navab, Nassir Tech Univ Munich Chair Comp Aided Med Procedures & Augmented Real C Munich Germany Huawei Munich Res Ctr D-80992 Munich Germany Munich Ctr Machine Learning MCML Munich Germany Tech Univ Munich D-80333 Munich Germany VisualAIs Labs GmbH Munich Germany Univ Stuttgart Inst Photogrammetry Stuttgart Germany

The generation of high-fidelity view synthesis is essential for robotic navigation and interaction but remains challenging, particularly in indoor environments and real-time scenarios. Existing techniques often require significant computational resources for both training and rendering, and they frequently result in suboptimal 3D representations due to insufficient geometric structuring. To address these limitations, we introduce VoxNeRF, a novel approach that utilizes easy-to-obtain geometry priors to enhance both the quality and efficiency of neural indoor reconstruction and novel view synthesis. We propose an efficient voxel-guided sampling technique that allocates computational resources selectively to the most relevant segments of rays based on a voxel-encoded geometry prior, significantly reducing training and rendering time. Additionally, we incorporate a robust depth loss to improve reconstruction and rendering quality in sparse view settings. Our approach is validated with extensive experiments on ScanNet and ScanNet++ where VoxNeRF outperforms existing state-of-the-art methods and establishes a new benchmark for indoor immersive interpolation and extrapolation settings.

关键词： Geometry Neural radiance field Rendering (computer graphics) Three-dimensional displays Image reconstruction Training Surface reconstruction Noise Image color analysis Solid modeling Visual learning vision-based navigation novel view synthesis indoor scene reconstruction

来源：评论

学校读者我要写书评

暂无评论

Audio-Based Kinship Verification Using Age Domain Conversion

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 301-305页

作者： Sun, Qiyang Akman, Alican Jing, Xin Milling, Manuel Schuller, Bjorn W. Imperial Coll London Dept Comp GLAM London SW7 2AZ England Tech Univ Munich CHI Chair Hlth Informat MRI D-81675 Munich Germany MDSI Munich Data Sci Inst D-85748 Munich Germany MCML Munich Ctr Machine Learning D-80539 Munich Germany

Audio-based kinship verification (AKV) is important in many domains, such as home security monitoring, forensic identification, and social network analysis. A key challenge in the task arises from differences in age across samples from different individuals, which can be interpreted as a domain bias in a cross-domain verification task. To address this issue, we design the notion of an "age-standardised domain" wherein we utilise the optimised CycleGAN-VC3 network to perform age-audio conversion to generate the in-domain audio. The generated audio dataset is employed to extract a range of features, which are then fed into a metric learning architecture to verify kinship. Experiments are conducted on the KAN_AV audio *** results demonstrate that the method markedly enhances the accuracy of kinship verification, while also offering novel insights for future kinship verification research.

关键词： Feature extraction Training Generators Videos Generative adversarial networks Vectors Hidden Markov models Data models Time-frequency analysis Recording Audio kinship verification deep learning GAN voice conversion

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：