检索结果-内蒙古大学图书馆

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

来源：评论

学校读者我要写书评

暂无评论

xMIL: Insightful Explanations for Multiple Instance learning in Histopathology 38

xMIL: Insightful Explanations for Multiple Instance Learning...

引用

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：

来源：评论

学校读者我要写书评

暂无评论

Improving Generalization and Convergence by Enhancing Implicit Regularization

arXiv

引用

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

来源：评论

学校读者我要写书评

暂无评论

UNSUPERVISED CROSS-CORPUS SPEECH EMOTION RECOGNITION USING DOMAIN-ADAPTIVE SUBSPACE learning

UNSUPERVISED CROSS-CORPUS SPEECH EMOTION RECOGNITION USING D...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Na Liu Yuan Zong Baofeng Zhang Li Liu Jie Chen Guoying Zhao Junchao Zhu School of Computer Science and Engineering Tianjin University of Technology China Research Center for Learning Science Southeast University China School of Electrical and Electronic Engineering Tianjin University of Technology China Center for Machine Vision and Signal Analysis University of Oulu Finland

ISBN: (纸本)9781538646595

In this paper, we investigate an interesting problem, i.e., unsupervised cross-corpus speech emotion recognition (SER), in which the training and testing speech signals come from two different speech emotion corpora. Meanwhile, the training speech signals are labeled, while the label information of the testing speech signals is entirely unknown. Due to this setting, the training (source) and testing (target) speech signals may have different feature distributions and therefore lots of existing SER methods would not work. To deal with this problem, we propose a domain-adaptive subspace learning (DoSL) method for learning a projection matrix with which we can transform the source and target speech signals from the original feature space to the label space. The transformed source and target speech signals in the label space would have similar feature distributions. Consequently, the classifier learned on the labeled source speech signals can effectively predict the emotional states of the unlabeled target speech signals. To evaluate the performance of the proposed DoSL method, we carry out extensive cross-corpus SER experiments on three speech emotion corpora including EmoDB, eNTERFACE, and AFEW 4.0. Compared with recent state-of-the-art cross-corpus SER methods, the proposed DoSL can achieve more satisfactory overall results.

关键词： Cross-corpus evaluation speech emotion recognition domain adaptation subspace learning

来源：评论

学校读者我要写书评

暂无评论

Portable tracker for neurophysiological research of sport shooting

Portable tracker for neurophysiological research of sport sh...

引用

Saratov Fall Meeting 2021: Computational Biophysics and Nanobiophotonics

作者： Antipov, V.M. Badarin, A.A. Grubov, V.V. Kazantsev, V.B. Hramov, A.E. Neuroscience and Cognivite Technology Laboratory Center for Technologies in Robotics and Mechatronics Components Innopolis University Universitetskaya Str. 1 Innopolis 420500 Russia Laboratory of Advanced Methods for High-Dimensional Data Analysis Lobachevsky State University of Nizhni Novgorod 23 Gagarin ave. Nizhny Novgorod603950 Russia Center for Neurotechnology and Machine Learning Immanuel Kant Baltic Federal University A. Nevskogo ul. 14 Kaliningrad236016 Russia Neurotechnology Deparment Lobachevsky State University of Nizhny Novgorod Nizhny Novgorod603022 Russia

ISBN: (数字)9781510653702

ISBN: (纸本)9781510653696

In this work we present the development process of a wireless portable module. It is developed to record various characteristics during sport shooting, such as automatic detection of the moment of shot and barrel movement during aiming, taking into account the peculiarities of its use in neurophysiological research. We propose an approach allowing to synchronize devices in a wireless local network with high accuracy (synchronization accuracy was 2 ms), as well as a method of logging the recorded data at a sampling rate of up to 2 kHz on an onboard flash drive. © 2022 SPIE.

关键词： Neurophysiology

来源：评论

学校读者我要写书评

暂无评论

Super Wide Regression Network for Unsupervised Cross-database Facial Expression Recognition

Super Wide Regression Network for Unsupervised Cross-Databas...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Na Liu Baofeng Zhang Yuan Zong Li Liu Jie Chen Guoying Zhao Lunchao Zhu School of Computer Science and Engineering Tianjin University of Technology China School of Electrical and Electronic Engineering Tianjin University of Technology China Research Center for Learning Science Southeast University China Center for Machine Vision and Signal Analysis University of Oulu Finland

ISBN: (纸本)9781538646595

Unsupervised cross-database facial expression recognition (FER) is a challenging problem, in which the training and testing samples belong to different facial expression databases. For this reason, the training (source) and testing (target) facial expression samples would have different feature distributions and hence the performance of lots of existing FER methods may decrease. To solve this problem, in this paper we propose a novel super wide regression network (SWiRN) model, which serves as the regression parameter to bridge the original feature space and the label space and herein in each layer the maximum mean discrepancy (MMD) criterion is used to enforce the source and target facial expression samples to share the same or similar feature distributions. Consequently, the learned SWiRN is able to predict the expression categories of the target samples although we have no access to any label information of target samples. We conduct extensive cross-database FER experiments on CK+, eNTERFACE, and Oulu-CASIA VIS facial expression databases to evaluate the proposed SWiRN. Experimental results show that our SWiRN model achieves more promising performance than recent proposed cross-database emotion recognition methods.

关键词： Cross-database facial expression recognition transfer learning domain adaptation super wide network Transfer (Psychology) Facial Expression label save area FER Gene Iron Eigenvector

来源：评论

学校读者我要写书评

暂无评论

data centric domain adaptation for historical text with OCR errors

arXiv

引用

arXiv 2021年

作者： März, Luisa Schweter, Stefan Poerner, Nina Roth, Benjamin Schütze, Hinrich Center for Information and Language Processing Ludwig Maximilian University Munich Germany Digital Philology Research Group Data Mining and Machine Learning University of Vienna Austria Bayerische Staatsbibliothek München Digital Library/Munich Digitization Center Munich Germany NLP Expert Center Data:Lab Volkswagen AG Munich Germany

We propose new methods for in-domain and cross-domain Named Entity Recognition (NER) on historical data for Dutch and French. For the cross-domain case, we address domain shift by integrating unsupervised in-domain data via contextualized string embeddings;and OCR errors by injecting synthetic OCR errors into the source domain and address data centric domain adaptation. We propose a general approach to imitate OCR errors in arbitrary input data. Our cross-domain as well as our in-domain results outperform several strong baselines and establish state-of-the-art results. We publish preprocessed versions of the French and Dutch Europeana NER corpora. © 2021, CC BY.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

learning trivializing gradient flows for lattice gauge theories

引用

Physical Review D 2023年第5期107卷 L051504-L051504页

作者： Simone Bacchio Pan Kessel Stefan Schaefer Lorenz Vaitl Computation-based Science and Technology Research Center The Cyprus Institute Nicosia Cyprus Machine Learning Group Technische Universität Berlin Berlin Germany BIFOLD Berlin Institute for the Foundations of Learning and Data Berlin Germany John von Neumann-Institut für Computing NIC Deutsches Elektronen-Synchrotron DESY Germany

We propose a unifying approach that starts from the perturbative construction of trivializing maps by Lüscher and then improves on it by learning. The resulting continuous normalizing flow model can be implemented using common tools of lattice field theory and requires several orders of magnitude fewer parameters than any existing machine learning approach. Specifically, our model can achieve competitive performance with as few as 14 parameters while existing deep-learning models have around 1 million parameters for SU(3) Yang-Mills theory on a 162 lattice. This has obvious consequences for training speed and interpretability. It also provides a plausible path for scaling machine-learning approaches toward realistic theories.

关键词： Lattice gauge theory Metropolis algorithm Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

Modeling the dielectric constant of silicon-based nanocomposites using machine learning

Modeling the dielectric constant of silicon-based nanocompos...

引用

2020 International Conference on Actual Problems of Electron Devices Engineering, APEDE 2020

作者： Korchagin, Sergey Alekseevich Klinaev, Yuri Vasilievich Terin, Denis Vladimirovich Romanchuk, Sergey Petrovich Financial University Government of the Russian Federation Department of Data Analysis and Machine Learning Moscow Russia Yuri Gagarin Saratov State Technical University Department of Natural and Mathematical Sciences Engels Russia Technol. and Qual. Mgmt. Saratov National Research State University Named after N. G. Chernyshevsky Department of Materials Science Saratov Russia Yuri Gagarin Saratov State Technical University Department of Information Security of Automated Systems Saratov Russia

ISBN: (数字)9781728143095

ISBN: (纸本)9781728143088

In this work, we solve the problem of predicting the dielectric constant of silicon-based nanocomposites using machine learning methods. Mathematical models and programs have been developed to predict the electrophysical properties of new materials, as well as to select the optimal composition. The results obtained will be useful for faster and cheaper creation of new functional composites. © 2020 IEEE.

关键词： Nanocomposites

来源：评论

学校读者我要写书评

暂无评论

RecycleNet: Latent Feature Recycling Leads to Iterative Decision Refinement

arXiv

引用

arXiv 2023年

作者： Koehler, Gregor Wald, Tassilo Ulrich, Constantin Zimmerer, David Jaeger, Paul F. Franke, Jörg K.H. Kohl, Simon Isensee, Fabian Maier-Hein, Klaus H. Heidelberg Division of Medical Image Computing Germany Helmholtz Information and Data Science School for Health Karlsruhe Heidelberg Germany Helmholtz Imaging DKFZ Germany NCT Heidelberg a partnership between DKFZ University Medical Center Heidelberg Germany Interactive Machine Learning Group DKFZ Applied Computer Vision Lab DKFZ Machine Learning Lab University of Freiburg Freiburg Germany London United Kingdom Pattern Analysis and Learning Group Heidelberg University Hospital Heidelberg Germany

Despite the remarkable success of deep learning systems over the last decade, a key difference still remains between neural network and human decision-making: As humans, we can not only form a decision on the spot, but also ponder, revisiting an initial guess from different angles, distilling relevant information, arriving at a better decision. Here, we propose RecycleNet, a latent feature recycling method, instilling the pondering capability for neural networks to refine initial decisions over a number of recycling steps, where outputs are fed back into earlier network layers in an iterative fashion. This approach makes minimal assumptions about the neural network architecture and thus can be implemented in a wide variety of contexts. Using medical image segmentation as the evaluation environment, we show that latent feature recycling enables the network to iteratively refine initial predictions even beyond the iterations seen during training, converging towards an improved decision. We evaluate this across a variety of segmentation benchmarks and show consistent improvements even compared with top-performing segmentation methods. This allows trading increased computation time for improved performance, which can be beneficial, especially for safety-critical applications. © 2023, CC BY.

关键词： Network layers

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：