检索结果-内蒙古大学图书馆

On the Effectiveness of Heterogeneous Ensemble Methods for Re-Identification

学校读者我要写书评

暂无评论

On the Effectiveness of Heterogeneous Ensemble Methods for R...

International Conference on machine learning and Applications (ICMLA)

作者： Simon Klüttermann Jérôme Rutinowski Frederik Polachowski Anh Nguyen Britta Grimme Moritz Roidl Emmanuel Müller TU Dortmund University Dortmund Germany Lamarr Institute for Machine Learning and Artificial Intelligence Dortmund Germany Paderborn University Paderborn Germany Research Center Trustworthy Data Science and Security Dortmund Germany

ISBN: (数字)9798350374889

ISBN: (纸本)9798350374896

In this contribution, we introduce a novel ensemble method for the re-identification of industrial entities, using images of chipwood pallets and galvanized metal plates as dataset examples. Our algorithms replace commonly used, complex siamese neural networks with an ensemble of simplified, rudimentary models, providing wider applicability, especially in hardware-restricted scenarios. Each ensemble sub-model uses different types of extracted features of the given data as its input, allowing for the creation of effective ensembles in a fraction of the training duration needed for more complex state-of-the-art models. We reach state-of-the-art performance at our task, with a Rank-1 accuracy of over 77% and a Rank-10 accuracy of over 99%, and introduce five distinct feature extraction approaches, and study their combination using different ensemble methods.

关键词： Training Accuracy Neural networks Metals Feature extraction Silicon Ensemble learning Security Reliability Pallets

PAM: A Propagation-Based Model for Segmenting Any 3D Objects across Multi-Modal Medical Images

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Zifan Nan, Xinyu Li, Jiazheng Zhao, Jie Li, Haifeng Lin, Ziling Li, Haoshen Chen, Heyun Liu, Yiting Tang, Lei Zhang, Li Dong, Bin Center for Data Science Peking University Beijing China Department of Radiology Key Laboratory of Carcinogenesis and Translational Research Ministry of Education Peking University Cancer Hospital and Institute Beijing China National Engineering Laboratory for Big Data Analysis and Applications Peking University Beijing China Peking University Beijing China Center for Machine Learning Research Peking University Beijing China National Biomedical Imaging Center Peking University Beijing China

Background: Volumetric segmentation is crucial for medical imaging applications but faces significant challenges. Current approaches often require extensive manual annotations and scenario-specific model training, limiting their transferability across different tasks or modalities. While general segmentation models offer some versatility in natural image processing, they struggle with the unique characteristics of medical images. There is an urgent need in clinical practice for a new segmentation approach that can effectively handle medical imagery features while maintaining adaptability across various three-dimensional objects and imaging modalities. Methods: We introduce PAM (Propagating Anything Model), a propagation-based segmentation approach that operates on 3D medical image volumes using a 2D prompt (bounding box or sketch mask). PAM extrapolates this initial input to generate a complete 3D segmentation by modeling inter-slice structural relationships, establishing a continuous information flow within 3D medical structures. This approach enhances segmentation effectiveness across various imaging modalities by focusing on structural and semantic continuities rather than isolating specific objects. The model combines a CNN-based UNet architecture for intra-slice information processing with a Transformer-based attention module to facilitate inter-slice propagation. This innovative framework results in a method with unique generalizability, capable of segmenting diverse 3D objects across different medical imaging modalities. Results: PAM demonstrated superior performance on 44 diverse medical datasets, notably improving the dice similarity coefficient (DSC) for hundreds of segmentation object types and various medical imaging modalities. Compared to modern models like MedSAM and SegVol, PAM achieved an average DSC improvement of over 18.1%, while maintaining stable predictions despite prompt deviation (one-way ANOVA test, P ≥ 0.5985) and varying propagation confi

关键词： Semantic Segmentation

xMIL: Insightful Explanations for Multiple Instance learning in Histopathology 38

学校读者我要写书评

暂无评论

xMIL: Insightful Explanations for Multiple Instance Learning...

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Hense, Julius Idaji, Mina Jamshidi Eberle, Oliver Schnake, Thomas Dippel, Jonas Ciernik, Laure Buchstab, Oliver Mock, Andreas Klauschen, Frederick Müller, Klaus-Robert Berlin Institute for the Foundations of Learning and Data Berlin Germany Machine Learning Group Technische Universität Berlin Berlin Germany Aignostics GmbH Berlin Germany Institute of Pathology Ludwig Maximilian University Munich Germany German Cancer Research Center Heidelberg Germany German Cancer Consortium Munich Germany Institute of Pathology Charité Universitätsmedizin Berlin Germany Department of Artificial Intelligence Korea University Seoul Korea Republic of Max-Planck Institute for Informatics Saarbrücken Germany

Multiple instance learning (MIL) is an effective and widely used approach for weakly supervised machine learning. In histopathology, MIL models have achieved remarkable success in tasks like tumor detection, biomarker prediction, and outcome prognostication. However, MIL explanation methods are still lagging behind, as they are limited to small bag sizes or disregard instance interactions. We revisit MIL through the lens of explainable AI (XAI) and introduce xMIL, a refined framework with more general assumptions. We demonstrate how to obtain improved MIL explanations using layer-wise relevance propagation (LRP) and conduct extensive evaluation experiments on three toy settings and four real-world histopathology datasets. Our approach consistently outperforms previous explanation attempts with particularly improved faithfulness scores on challenging biomarker prediction tasks. Finally, we showcase how xMIL explanations enable pathologists to extract insights from MIL models, representing a significant advance for knowledge discovery and model debugging in digital histopathology. Codes are available at: https://***/bifold-pathomics/xMIL. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

Improving Generalization and Convergence by Enhancing Implicit Regularization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Mingze Wang, Jinbo He, Haotian Wang, Zilin Huang, Guanhua Xiong, Feiyu Li, Zhiyu Weinan, E. Wu, Lei School of Mathematical Sciences Peking University China Center for Machine Learning Research Peking University China China AI for Science Institute China School of Data Science University of Science and Technology of China China ByteDance Research China

In this work, we propose an Implicit Regularization Enhancement (IRE) framework to accelerate the discovery of flat solutions in deep learning, thereby improving generalization and convergence. Specifically, IRE decouples the dynamics of flat and sharp directions, which boosts the sharpness reduction along flat directions while maintaining the training stability in sharp directions. We show that IRE can be practically incorporated with generic base optimizers without introducing significant computational overload. Experiments show that IRE consistently improves the generalization performance for image classification tasks across a variety of benchmark datasets (CIFAR-10/100, ImageNet) and models (ResNets and ViTs). Surprisingly, IRE also achieves a 2× speed-up compared to AdamW in the pre-training of Llama models (of sizes ranging from 60M to 229M) on datasets including Wikitext-103, Minipile, and Openwebtext. Moreover, we provide theoretical guarantees, showing that IRE can substantially accelerate the convergence towards flat minima in sharpness-aware minimization (SAM). Copyright © 2024, The Authors. All rights reserved.

关键词： Image classification

Using Convolutional Neural Network to Classify 2D EEG Scalp Topograms during Visual Task 5

学校读者我要写书评

暂无评论

Using Convolutional Neural Network to Classify 2D EEG Scalp ...

5th Scientific School on Dynamics of Complex Networks and their Applications, DCNA 2021

作者： Korchagin, Sergey Maksimenko, Vladimir Hramov, Alexander Department of Data Analysis and Machine Learning Financial University under the Government of the Russian Federation Moscow Russia Institute of Information Technologies Mathematics and Mechanics Lobachevsky State University of Nizhny Novgorod Nizhny Novgorod Russia Center for Neurotechnology and Machine Learning Immanuel Kant Baltic Federal University Kaliningrad Russia

ISBN: (纸本)9781665442824

we developed an artificial neural network (ANN) classifier to analyze the cortical activity signals during visual information processing. We tested several ANN architectures and chose a convolutional neural network (CNN) with the Resnet50 topology. As a result, the trained CNN classifier achieved an accuracy of 74% on the data of newly recruited subjects. © 2021 IEEE

关键词： Convolutional neural networks

Stroke Home Rehabilitation Approach Using Mobile Application Based on PostNet machine learning Model 23

学校读者我要写书评

暂无评论

Stroke Home Rehabilitation Approach Using Mobile Application...

7th International Conference on Medical and Health Informatics, ICMHI 2023

作者： Das, Utpal Chandra Le, Ngoc Thien Benjapolakul, Watit Vitoonpong, Timporn Pluempitiwiriyawej, Charnchai Center of Excellence in Artificial Intelligence Machine Learning and Smart Grid Technology Department of Electrical Engineering Chulalongkorn University Bangkok10330 Thailand Department of Rehabilitation Medicine Faculty of Medicine Chulalongkorn University Bangkok10330 Thailand Multimedia Data Analytics and Processing Research Unit Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok10330 Thailand

ISBN: (纸本)9798400700712

Stroke is a significant cause of mortality and disability globally, with its occurrence in the human brain and motor function being linked to various parts of the human body. Stroke victims often experience disabilities or mobility problems in affected body parts, either on one or both sides of the body. Physiotherapy exercises are the primary treatment and medication for stroke patients, necessitating daily monitoring by a physiotherapist. However, this approach is expensive, and rehabilitation centers and physicians are scarce. Numerous research studies have been conducted to address this issue from various perspectives. This study proposes a Convolutional Neural Network (CNN) based pose net machine learning (ML) model for stroke home rehabilitation using pose detection and classification with a skeleton base model and human pose estimation drawing. We trained our ML model using different human pose images for this model. The tested accuracy of our CNN model is 100% for our exercise pose image test case. Later we build and mobile application for remote rehabilitation. Tested the application with ten different subjects, achieving a 98% for elbow extension, and in the case of elbow flexion and Normal position of balancing both sides, all the persons achieved a 100 % accuracy rate under a laboratory environment. © 2023 ACM.

关键词： Patient treatment

Stochastic Principles of Optimality in Games with Nature and Their Application in Investment Management 14

学校读者我要写书评

暂无评论

Stochastic Principles of Optimality in Games with Nature and...

14th International Conference Management of Large-Scale System Development, MLSD 2021

作者： Gorelik, Victor Zolotova, Tatiana Moscow Pedagogical State University Department of Simulation Systems and Operations Research FRC CSC RAS Moscow Russia Financial University Government of the Russian Federation Department of Data Analysis and Machine Learning Moscow Russia

ISBN: (纸本)9781665412308

The principle of optimality in decision-making for games with nature, based on assessments of efficiency and risk, is proposed. In contrast to the traditional approach to the definition of a mixed strategy in game theory, this work considers the possibility of a correlation dependence of random values of payoffs for the original alternatives. © 2021 IEEE.

关键词： Game theory

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