检索结果-内蒙古大学图书馆

A CALIBRATION TEST FOR EVALUATING SET-BASED EPISTEMIC UNCERTAINTY REPRESENTATIONS

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jürgens, Mira Mortier, Thomas Hüllermeier, Eyke Bengs, Viktor Waegeman, Willem Department of Data Analysis and Mathematical Modeling Ghent University Belgium Department of Environment Ghent University Belgium Department of Informatics Munich Center for Machine Learning LMU Munich Germany

The accurate representation of epistemic uncertainty is a challenging yet essential task in machine learning. A widely used representation corresponds to convex sets of probabilistic predictors, also known as credal sets. One popular way of constructing these credal sets is via ensembling or specialized supervised learning methods, where the epistemic uncertainty can be quantified through measures such as the set size or the disagreement among members. In principle, these sets should contain the true data-generating distribution. As a necessary condition for this validity, we adopt the strongest notion of calibration as a proxy. Concretely, we propose a novel statistical test to determine whether there is a convex combination of the set’s predictions that is calibrated in distribution. In contrast to previous methods, our framework allows the convex combination to be instance dependent, recognizing that different ensemble members may be better calibrated in different regions of the input space. Moreover, we learn this combination via proper scoring rules, which inherently optimize for calibration. Building on differentiable, kernel-based estimators of calibration errors, we introduce a nonparametric testing procedure and demonstrate the benefits of capturing instance-level variability on of synthetic and real-world experiments. © 2025, CC BY.

关键词： Self-supervised learning

A MEMORY EFFICIENT RANDOMIZED SUBSPACE OPTIMIZATION METHOD FOR TRAINING LARGE LANGUAGE MODELS

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Chen, Yiming Zhang, Yuan Liu, Yin Yuan, Kun Wen, Zaiwen Beijing International Center for Mathematical Research Peking University Beijing China Center for Data Science Peking University Beijing China Center for Machine Learning Research Peking University Beijing China

The memory challenges associated with training Large Language Models (LLMs) have become a critical concern, particularly when using the Adam optimizer. To address this issue, numerous memory-efficient techniques have been proposed, with GaLore standing out as a notable example designed to reduce the memory footprint of optimizer states. However, these approaches do not alleviate the memory burden imposed by activations, rendering them unsuitable for scenarios involving long context sequences or large mini-batches. Moreover, their convergence properties are still not well-understood in the literature. In this work, we introduce a Randomized Subspace Optimization framework for pre-training and fine-tuning LLMs. Our approach decomposes the high-dimensional training problem into a series of lower-dimensional subproblems. At each iteration, a random subspace is selected, and the parameters within that subspace are optimized. This structured reduction in dimensionality allows our method to simultaneously reduce memory usage for both activations and optimizer states. We establish comprehensive convergence guarantees and derive rates for various scenarios, accommodating different optimization strategies to solve the subproblems. Extensive experiments validate the superior memory and communication efficiency of our method, achieving performance comparable to GaLore and Adam. Copyright © 2025, The Authors. All rights reserved.

关键词：

U-Star: AN Asymmetric U-Shaped Network Based on Element-Wise Multiplication to Segment Nuclei in H&E Stained Histological Images

学校读者我要写书评

暂无评论

U-Star: AN Asymmetric U-Shaped Network Based on Element-Wise...

IEEE International Symposium on Biomedical Imaging

作者： Guangzhengao Yang Li Zhang Jie Zhao Zifan Chen Haoshen Li Bin Dong Center for Data Science Peking University China National Engineering Laboratory for Big Data Analysis and Applications Peking University China Peking University Changsha Institute for Computing and Digital Economy China Beijing International Center for Mathematical Research Peking University China Center for Machine Learning Research Peking University China National Biomedical Imaging Center Peking University China

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

Nuclei segmentation in Hematoxylin and Eosin (H&E) stained images plays a crucial role in cancer diagnosis and pathological evaluation, enabling pathologists to identify abnormal cells and assess their morphology and distribution. While current automated nuclei segmentation methods predominantly employ convolutional neural networks and attention mechanisms, the potential of element-wise multiplication has been largely unexplored. This paper introduces U-Star, a novel asymmetric segmentation network based on the star block that leverages element-wise multiplication. U-Star adopts the classic encoder-decoder architecture and innovatively implements star-connection as an alternative to traditional skip-connections. In experiments on an H&E stained image dataset, U-Star achieved superior performance with a Dice coefficient of 0.8783, accuracy of 0.9089, and IoU of 0.7929, significantly outperforming baseline models. Extensive ablation studies validate the effectiveness of the star-connection and demonstrate the advantages of our proposed framework. Beyond advancing the application of element-wise multiplication techniques, U-Star shows promising potential for broader applications in medical image segmentation.

关键词： Image segmentation Pathology Technological innovation Image analysis Stars Morphology Network architecture Medical diagnostic imaging Tuning Cancer

Multi-Stage Bidirectional Cross-Attention Model for Predicting Prognosis in Multiple Peritoneum Lesions with Clinical Information

学校读者我要写书评

暂无评论

Multi-Stage Bidirectional Cross-Attention Model for Predicti...

IEEE International Symposium on Biomedical Imaging

作者： Haoshen Li Jieyuan Cai Yiyuan Wei Zifan Chen Heyun Chen Jie Zhao Yanjie Shi Bin Dong Lei Tang Xiaotian Zhang Li Zhang Center for Data Science Peking University China Peking University Cancer Hospital & Institute China National Engineering Laboratory for Big Data Analysis and Applications Peking University China Peking University Changsha Institute for Computing and Digital Economy China Beijing International Center for Mathematical Research Peking University China Center for Machine Learning Research Peking University China

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

Peritoneal metastasis occurs when cancer cells spread from the primary tumor to the peritoneum, leading to morphological alterations that significantly impact patient survival. The specific changes across multiple peritoneal sites can effectively indicate a patient's risk level and prognosis. However, the considerable variation in peritoneal shape, size, and location poses significant challenges for feature extraction and prognostic analysis. Traditional multi-instance learning approaches typically fuse instance features extracted by the backbone network at the final stage, but their lack of intermediate interaction limits the ability to capture both common and distinct features. To address this limitation, we propose a Multi-Stage BiDirectional Cross-Attention framework (MSBDCA). Our approach enhances feature extraction and prognostic analysis by facilitating interaction among different peritoneal instances and incorporating clinical information. Experimental comparisons with five baseline models demonstrate our method's superiority, achieving improvements of 8.7% in Area Under Curve (AUC) and 5.9% in Concordance Index (C_index). These results suggest a promising direction for survival and prognostic analysis using peritoneal lesions. Our code is available at https://***/HaoshenLi/MSBDCA.

关键词： learning systems Shape Predictive models Feature extraction Metastasis Lesions Clinical diagnosis Prognostics and health management Abdomen Biomedical imaging

DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Aids

学校读者我要写书评

暂无评论

DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Ai...

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Tsangko, Iosif Triantafyllopoulos, Andreas Müller, Michael Schröter, Hendrik Schuller, Björn W. EIHW - Embedded Intelligence for Health Care and Wellbeing University of Augsburg Germany Technical University of Munich Germany MCML - Munich Center for Machine Learning Munich Germany WS Audiology Research and Development Erlangen Germany GLAM - Group on Language Audio & Music Imperial College London United Kingdom MDSI - Munich Data Science Institute Munich Germany

ISBN: (纸本)9798350368741

The DeepFilterNet (DFN) architecture was recently proposed as a deep learning model suited for hearing aid devices. Despite its competitive performance on numerous benchmarks, it still follows a 'one-size-fits-all' approach, which aims to train a single, monolithic architecture that generalises across different noises and environments. However, its limited size and computation budget can hamper its generalisability. Recent work has shown that in-context adaptation can improve performance by conditioning the denoising process on additional information extracted from background recordings to mitigate this. These recordings can be offloaded outside the hearing aid, thus improving performance while adding minimal computational overhead. We introduce these principles to the DFN model, thus proposing the DFingerNet (DFiN) model, which shows superior performance on various benchmarks inspired by the DNS Challenge. © 2025 IEEE.

关键词： Context Adaptation Deep learning Hearing Aids Speech Denoising Speech Enhancement

Conformal Prediction in Hierarchical Classification

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Mortier, Thomas Javanmardi, Alireza Sale, Yusuf Hüllermeier, Eyke Waegeman, Willem Department of Environment Ghent University Ghent Belgium Department of Data Analysis and Mathematical Modelling Ghent University Ghent Belgium Institute of Informatics LMU Munich Munich Germany Munich Center for Machine Learning Munich Germany

Conformal prediction has emerged as a widely used framework for constructing valid prediction sets in classification and regression tasks. In this work, we extend the split conformal prediction framework to hierarchical classification, where prediction sets are commonly restricted to internal nodes of a predefined hierarchy, and propose two computationally efficient inference algorithms. The first algorithm returns internal nodes as prediction sets, while the second relaxes this restriction, using the notion of representation complexity, yielding a more general and combinatorial inference problem, but smaller set sizes. Empirical evaluations on several benchmark datasets demonstrate the effectiveness of the proposed algorithms in achieving nominal coverage. © 2025, CC BY.

关键词： Prediction models

RVPD: An Automated System for Calculating the Tortuosity and Bifurcation Angles of Retinal Vessels to Predict Diseases

学校读者我要写书评

暂无评论

RVPD: An Automated System for Calculating the Tortuosity and...

IEEE International Symposium on Biomedical Imaging

作者： Guangzhengao Yang Xingyu Luo Jie Zhao Fangfang Fan Haoshen Li Bin Dong Li Zhang Yan Zhang Center for Data Science Peking University China Department of Cardiology Peking University First Hospital China National Engineering Laboratory for Big Data Analysis and Applications Peking University China Peking University Changsha Institute for Computing and Digital Economy China Beijing International Center for Mathematical Research Peking University China Center for Machine Learning Research Peking University China National Biomedical Imaging Center Peking University China

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

Hypertension and diabetes are known to potentially cause morphological changes in the retinal capillary system, yet quantifying these changes presents significant challenges. This research addresses this issue by designing and developing an automated system capable of analyzing retinal images to detect and measure these changes. Our system automatically segments vessels and the optic disc, calculates tortuosity and bifurcation angles, and performs disease prediction based on these measurements along with clinical features. For tortuosity calculation, we adopted a novel curve fitting method, which provides more stable convergence compared to traditional discrete numerical methods. The system integrates clinical data and conducts prediction experiments on hypertensive and diabetic patients with varying sample sizes. Our results demonstrate the system's strong performance and scalability: in the hypertension group, the AUC reached a maximum of 0.814 ± 0.023, with an ACC of 0.735 ± 0.033, while in the diabetes group, the AUC reached a maximum of 0.710 ± 0.043, with an ACC of 0.641 ± 0.052. These findings suggest that our automated system could serve as a valuable tool for early detection and risk assessment of hypertension and diabetes, potentially improving patient outcomes through timely intervention.

关键词： Hypertension Image segmentation Scalability Bifurcation Optical variables measurement Optical imaging Retinal vessels Diabetes Risk management Diseases

Nonequilibrium Universality of Rydberg-Excitation Spreading on a Dynamic Network

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ohler, Simon Brady, Daniel Mischke, Patrick Bender, Jana Ott, Herwig Niederprüm, Thomas Ripken, Winfried Otterbach, Johannes S. Fleischhauer, Michael Department of Physics Research Center OPTIMAS RPTU Kaiserslautern Kaiserslautern67663 Germany Mainz55128 Germany Machine Learning Group Technische Universität Berlin Berlin10587 Germany Berlin Institute for the Foundations of Learning and Data Berlin10587 Germany Orthogonal Otter UG Berlin10961 Germany

Understanding the universal properties of non-equilibrium phase transitions of spreading processes is a challenging problem. This applies in particular to irregular and dynamically varying networks. We here investigate an experimentally accessible model system for such processes, namely the absorbing-state phase transition (ASPT) of Rydberg-excitation spreading, known as Rydberg facilitation, in a laser-driven gas of mobile atoms. It occurs on an irregular graph, set by the random atom positions in the gas and, depending on temperature, changes its character from static to dynamic. By studying the behavior of the order parameter in [Phys. Rev. Lett. 133, 173401 (2024)] we showed numerical evidence for a crossover from directed percolation (DP) universality through various phases of anomalous directed percolation (ADP) to mean-field (MF) behavior when the temperature of the gas is increased. As the behavior of the order parameter is not sufficient to uniquely determine the universality class, we here analyze the distribution of avalanches – characteristic of non-equilibrium critical behavior – to fully characterize the ASPT. Performing extended numerical calculations and experiments on a cold 87Rb atom gas we confirm our earlier numerical findings and our phenomenological model that maps the dynamic network to a static one with power-law tails of the distribution of excitation distances. Furthermore we discuss the influence of dissipation, present in the experiment and a necessary ingredient for the self-organization of the system to the critical point. In particular we study the potential modification of the universality class by losses as a function of dissipation strength. © 2025, CC BY.

关键词： Rydberg states

GraphMorph: Tubular Structure Extraction by Morphing Predicted Graphs

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhang, Zhao Zhao, Ziwei Wang, Dong Wang, Liwei Center for Data Science Peking University China Yizhun Medical AI Co. Ltd China State Key Laboratory of General Artificial Intelligence School of Intelligence Science and Technology Peking University China Center for Machine Learning Research Peking University China China

Accurately restoring topology is both challenging and crucial in tubular structure extraction tasks, such as blood vessel segmentation and road network extraction. Diverging from traditional approaches based on pixel-level classification, our proposed method, named GraphMorph, focuses on branch-level features of tubular structures to achieve more topologically accurate predictions. GraphMorph comprises two main components: a Graph Decoder and a Morph Module. Utilizing multi-scale features extracted from an image patch by the segmentation network, the Graph Decoder facilitates the learning of branch-level features and generates a graph that accurately represents the tubular structure in this patch. The Morph Module processes two primary inputs: the graph and the centerline probability map, provided by the Graph Decoder and the segmentation network, respectively. Employing a novel SkeletonDijkstra algorithm, the Morph Module produces a centerline mask that aligns with the predicted graph. Furthermore, we observe that employing centerline masks predicted by GraphMorph significantly reduces false positives in the segmentation task, which is achieved by a simple yet effective post-processing strategy. The efficacy of our method in the centerline extraction and segmentation tasks has been substantiated through experimental evaluations across various datasets. Source code will be released soon. Copyright © 2025, The Authors. All rights reserved.

关键词： Image segmentation