检索结果-内蒙古大学图书馆

arXiv 2023年

作者： He, Chunming Li, Kai Xu, Guoxia Yan, Jiangpeng Tang, Longxiang Zhang, Yulun Li, Xiu Wang, Yaowei Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen518055 China Machine Learning Department NEC Laboratories America Inc. NJ08540 United States Department of Computer Science Norwegian University of Science and Technology Gjovik2815 Norway The Computer Vision Lab ETH Zürich Zürich8092 Switzerland Peng Cheng Laboratory Shenzhen518066 China

Unpaired Medical Image Enhancement (UMIE) aims to transform a low-quality (LQ) medical image into a high-quality (HQ) one without relying on paired images for training. While most existing approaches are based on Pix2Pix/CycleGAN and are effective to some extent, they fail to explicitly use HQ information to guide the enhancement process, which can lead to undesired artifacts and structural distortions. In this paper, we propose a novel UMIE approach that avoids the above limitation of existing methods by directly encoding HQ cues into the LQ enhancement process in a variational fashion and thus model the UMIE task under the joint distribution between the LQ and HQ domains. Specifically, we extract features from an HQ image and explicitly insert the features, which are expected to encode HQ cues, into the enhancement network to guide the LQ enhancement with the variational normalization module. We train the enhancement network adversarially with a discriminator to ensure the generated HQ image falls into the HQ domain. We further propose a content-aware loss to guide the enhancement process with wavelet-based pixel-level and multi-encoder-based feature-level constraints. Additionally, as a key motivation for performing image enhancement is to make the enhanced images serve better for downstream tasks, we propose a bi-level learning scheme to optimize the UMIE task and downstream tasks cooperatively, helping generate HQ images both visually appealing and favorable for downstream tasks. Experiments on three medical datasets, including two newly collected datasets, verify that the proposed method outperforms existing techniques in terms of both enhancement quality and downstream task performance. We will make the code and the newly collected datasets publicly available for community study. Copyright © 2023, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Two-Dimensional Weisfeiler-Lehman Graph Neural Networks for Link Prediction

arXiv

引用

arXiv 2022年

作者： Hu, Yang Wang, Xiyuan Lin, Zhouchen Li, Pan Zhang, Muhan Institute for Artificial Intelligence Peking University China Key Lab. of Machine Perception School of Artificial Intelligence Peking University China Department of Computer Science Purdue University United States Beijing Institute for General Artificial Intelligence China

Link prediction is one important application of graph neural networks (GNNs). Most existing GNNs for link prediction are based on one-dimensional Weisfeiler-Lehman (1-WL) test. 1-WL-GNNs first compute node representations by iteratively passing neighboring node features to the center, and then obtain link representations by aggregating the pairwise node representations. As pointed out by previous works, this two-step procedure results in low discriminating power, as 1-WL-GNNs by nature learn node-level representations instead of link-level. In this paper, we study a completely different approach which can directly obtain node pair (link) representations based on two-dimensional Weisfeiler-Lehman (2-WL) tests. 2-WL tests directly use links (2-tuples) as message passing units instead of nodes, and thus can directly obtain link representations. We theoretically analyze the expressive power of 2-WL tests to discriminate non-isomorphic links, and prove their superior link discriminating power than 1-WL. Based on different 2-WL variants, we propose a series of novel 2-WL-GNN models for link prediction. Experiments on a wide range of real-world datasets demonstrate their competitive performance to state-of-the-art baselines and superiority over plain 1-WL-GNNs. Copyright © 2022, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

An empirical review of deep learning frameworks for change detection: Model design, experimental frameworks, challenges and research needs

arXiv

引用

arXiv 2021年

作者： Mandal, Murari Vipparthi, Santosh Kumar Vision Intelligence Lab Department of Computer Science and Engineering Malaviya National Instituteof Technology Jaipur302017 India

Visual change detection, aiming at segmentation of video frames into foreground and background regions, is one of the elementary tasks in computer vision and video analytics. The applications of change detection include anomaly detection, object tracking, traffic monitoring, human machine interaction, behavior analysis, action recognition, and visual surveillance. Some of the challenges in change detection include background fluctuations, illumination variation, weather changes, intermittent object motion, shadow, fast/slow object motion, camera motion, heterogeneous object shapes and real-time processing. Traditionally, this problem has been solved using hand-crafted features and background modelling techniques. In recent years, deep learning frameworks have been successfully adopted for robust change detection. This article aims to provide an empirical review of the state-of-the-art deep learning methods for change detection. More specifically, we present a detailed analysis of the technical characteristics of different model designs and experimental frameworks. We provide model design based categorization of the existing approaches, including the 2D-CNN, 3D-CNN, ConvLSTM, multi-scale features, residual connections, autoencoders and GAN based methods. Moreover, an empirical analysis of the evaluation settings adopted by the existing deep learning methods is presented. To the best of our knowledge, this is a first attempt to comparatively analyze the different evaluation frameworks used in the existing deep change detection methods. Finally, we point out the research needs, future directions and draw our own conclusions. © 2021, CC BY-NC-ND.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Dynamics and Computational Principles of Echo State Networks: A Mathematical Perspective

arXiv

引用

arXiv 2025年

作者： Singh, Pradeep Kumar, Ashutosh Ghosh, Sutirtha Hrishit, B.P. Raman, Balasubramanian Machine Intelligence Lab Department of Computer Science and Engineering Indian Institute of Technology Roorkee Roorkee247667 India

Reservoir computing (RC) represents a class of state-space models (SSMs) characterized by a fixed state transition mechanism (the reservoir) and a flexible readout layer that maps from the state space. It is a paradigm of computational dynamical systems that harnesses the transient dynamics of high-dimensional state spaces for efficient processing of temporal data. Rooted in concepts from recurrent neural networks, RC achieves exceptional computational power by decoupling the training of the dynamic reservoir from the linear readout layer, thereby circumventing the complexities of gradient-based optimization. This work presents a systematic exploration of RC, addressing its foundational properties such as the echo state property, fading memory, and reservoir capacity through the lens of dynamical systems theory. We formalize the interplay between input signals and reservoir states, demonstrating the conditions under which reservoirs exhibit stability and expressive power. Further, we delve into the computational trade-offs and robustness characteristics of RC architectures, extending the discussion to their applications in signal processing, time-series prediction, and control systems. The analysis is complemented by theoretical insights into optimization, training methodologies, and scalability, highlighting open challenges and potential directions for advancing the theoretical underpinnings of *** Codes 37N35, 37D45, 93C10, 93C35, 93C40, 93C55 © 2025, CC BY.

关键词： State space methods

来源：评论

学校读者我要写书评

暂无评论

Representation Matters for Mastering Chess: Improved Feature Representation in AlphaZero Outperforms Switching to Transformers

arXiv

引用

arXiv 2023年

作者： Czech, Johannes Blüml, Jannis Kersting, Kristian Steingrimsson, Hedinn Artificial Intelligence and Machine Learning Lab TU Darmstadt Germany Darmstadt Germany Centre for Cognitive Science TU Darmstadt Germany Darmstadt Germany Department of Electrical and Computer Engineering Rice University HoustonTX United States HoustonTX United States

While transformers have gained recognition as a versatile tool for artificial intelligence (AI), an unexplored challenge arises in the context of chess — a classical AI benchmark. Here, incorporating vision Transformers (ViTs) into AlphaZero is insufficient for chess mastery, mainly due to ViTs’ computational limitations. The attempt to optimize their efficiency by combining MobileNet and NextViT outperformed AlphaZero by about 30 Elo. However, we propose a practical improvement that involves a simple change in the input representation and value loss functions. As a result, we achieve a significant performance boost of up to 180 Elo points beyond what is currently achievable with AlphaZero in chess. In addition to these improvements, our experimental results using the Integrated Gradient technique confirm the effectiveness of the newly introduced features. © 2023, CC BY.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Deformable Symmetry Attention for Nuclear Medicine Image Segmentation

SSRN

引用

SSRN 2024年

作者： Zhang, Zeao Yang, Pei Liu, Ruomeng Cai, Huawei Zhao, Zhen Guo, Quan Yi, Zhang Machine Intelligence Lab College of Computer Science Sichuan University No. 24 Southern Section of First Ring Road Sichuan Chengdu610065 China Department of Nuclear Medicine West China Hospital of Sichuan University No. 17 Section 3 Renmin South Road Sichuan Chengdu610041 China

Prior knowledge of the medical domain has consistently enriched medical image analysis, yet its full potential remains to be explored. Our Deformable Symmetry Attention (DSA) aims to leverage anatomical symmetry prior, commonly used by clinicians, to enhance the automated analysis of medical images, especially in inherently grainy nuclear medicine images. DSA processes an explicit feature alignment attention. It starts with a prior alignment by reflecting the input image to leverage symmetry, followed by a fine-grained alignment using deformable image registration. Our novel Rotational Spatial Transformation Function (RSTF) enhances the registration by estimating local displacement with rotation in relative coordinates, enforcing local smoothness under anatomical variations. Additionally, the Attention Redistribution~(AR) module addresses pixel-level mismatch in grainy nuclear medicine images by a set of learnable Gaussian kernels. By translating the anatomical symmetry prior into a locality prior, DSA seamlessly integrates with and enhances sophisticated convolutional neural networks (CNNs) and transformer-based segmentation models. Evaluated on challenging adolescent Whole-Body Bone Scan (WBS) lesion segmentation as well as the extended 3D MRI brain tumor segmentation and chest X-ray anatomical segmentation, DSA consistently outperformed strong baseline methods, demonstrating its effectiveness and versatility across various experiment settings. It presents a novel and promising perspective for explicitly leveraging prior knowledge in medical image analysis. © 2024, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

CSP: Self-Supervised Contrastive Spatial Pre-Training for Geospatial-Visual Representations

arXiv

引用

arXiv 2023年

作者： Mai, Gengchen Lao, Ni He, Yutong Song, Jiaming Ermon, Stefano Spatially Explicit Artificial Intelligence Lab Department of Geography University of Georgia United States Department of Computer Science Stanford University United States School of Computing University of Georgia United States Google Inc United States Machine Learning Department Carnegie Mellon University United States

Geo-tagged images are publicly available in large quantities, whereas labels such as object classes are rather scarce and expensive to collect. Meanwhile, contrastive learning has achieved tremendous success in various natural image and language tasks with limited labeled data. However, existing methods fail to fully leverage geospatial information, which can be paramount to distinguishing objects that are visually similar. To directly leverage the abundant geospatial information associated with images in pre-training, fine-tuning, and inference stages, we present Contrastive Spatial Pre-Training (CSP), a self-supervised learning framework for geo-tagged images. We use a dual-encoder to separately encode the images and their corresponding geo-locations, and use contrastive objectives to learn effective location representations from images, which can be transferred to downstream supervised tasks such as image classification. Experiments show that CSP can improve model performance on both iNat2018 and fMoW dataset. Especially, on iNat2018, CSP significantly boosts the model performance with 10-34% relative improvement with various labeled training data sampling *** Codes 68T07, 68T45 © 2023, CC0.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Typology for Exploring the Mitigation of Shortcut Behavior

arXiv

引用

arXiv 2022年

作者： Friedrich, Felix Stammer, Wolfgang Schramowski, Patrick Kersting, Kristian Technical University of Darmstadt Computer Science Department Artificial Intelligence and Machine Learning Lab Darmstadt Germany Darmstadt Germany LAION Germany Germany Technical University of Darmstadt Centre for Cognitive Science Darmstadt Germany

As machine learning models become larger, and are increasingly trained on large and uncurated data sets in weakly supervised mode, it becomes important to establish mechanisms for inspecting, interacting, and revising models. These are necessary to mitigate shortcut learning effects and to guarantee that the model’s learned knowledge is aligned with human knowledge. Recently, several explanatory interactive machine learning (XIL) methods were developed for this purpose, but each have different motivations and methodological details. In this work, we provide a unification of various XIL methods into a single typology by establishing a common set of basic modules. We discuss benchmarks and other measures for evaluating the overall abilities of XIL methods. With this extensive toolbox, we systematically and quantitatively compare several XIL methods. In our evaluations, all methods are shown to improve machine learning models in terms of accuracy and explainability. However, we found remarkable differences in individual benchmark tasks, which reveal valuable application-relevant aspects for the integration of these benchmarks in the development of future methods.1 © 2022, CC BY.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Stability and Generalization Analysis of Gradient Methods for Shallow Neural Networks

arXiv

引用

arXiv 2022年

作者： Lei, Yunwen Jin, Rong Ying, Yiming School of Computer Science University of Birmingham United Kingdom Machine Intelligence Technology Lab Alibaba Group China Department of Mathematics and Statistics State University of New York Albany United States

While significant theoretical progress has been achieved, unveiling the generalization mystery of overparameterized neural networks still remains largely elusive. In this paper, we study the generalization behavior of shallow neural networks (SNNs) by leveraging the concept of algorithmic stability. We consider gradient descent (GD) and stochastic gradient descent (SGD) to train SNNs, for both of which we develop consistent excess risk bounds by balancing the optimization and generalization via early-stopping. As compared to existing analysis on GD, our new analysis requires a relaxed overparameterization assumption and also applies to SGD. The key for the improvement is a better estimation of the smallest eigenvalues of the Hessian matrices of the empirical risks and the loss function along the trajectories of GD and SGD by providing a refined estimation of their iterates. © 2022, CC BY.

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY labELS 9

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY L...

引用

9th International Conference on Learning Representations, ICLR 2021

作者： Xia, Xiaobo Liu, Tongliang Han, Bo Gong, Chen Wang, Nannan Ge, Zongyuan Chang, Yi Trustworthy Machine Learning Lab School of Computer Science The University of Sydney Australia Department of Computer Science Hong Kong Baptist University Hong Kong School of Computer Science and Engineering Nanjing University of Science and Technology China ISN State Key Laboratory School of Telecommunications Engineering Xidian University China Medical AI Group Faculty of Engineering Monash University Australia Airdoc Research Monash University Australia School of Artificial Intelligence Jilin University China

The memorization effects of deep networks show that they will first memorize training data with clean labels and then those with noisy labels. The early stopping method therefore can be exploited for learning with noisy labels. However, the side effect brought by noisy labels will influence the memorization of clean labels before early stopping. In this paper, motivated by the lottery ticket hypothesis which shows that only partial parameters are important for generalization, we find that only partial parameters are important for fitting clean labels and generalize well, which we term as critical parameters;while the other parameters tend to fit noisy labels and cannot generalize well, which we term as non-critical parameters. Based on this, we propose robust early-learning to reduce the side effect of noisy labels before early stopping and thus enhance the memorization of clean labels. Specifically, in each iteration, we divide all parameters into the critical and non-critical ones, and then perform different update rules for different types of parameters. Extensive experiments on benchmark-simulated and real-world label-noise datasets demonstrate the superiority of the proposed method over the state-of-the-art label-noise learning methods. © 2021 ICLR 2021 - 9th International Conference on Learning Representations. All rights reserved.

关键词： Iterative methods

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：