检索结果-内蒙古大学图书馆

Visualizing population structure with variational autoencoders

G3-GENES GENOMES GENETICS 2021年第1期11卷 jkaa036页

作者： Battey, C. J. Coffing, Gabrielle C. Kern, Andrew D. Univ Oregon Dept Biol Inst Ecol & Evolut Eugene OR 97403 USA

Dimensionality reduction is a common tool for visualization and inference of population structure from genotypes, but popular methods either return too many dimensions for easy plotting (PCA) or fail to preserve global geometry (t-SNE and UMAP). Here we explore the utility of variational autoencoders (VAEs)-generative machine learning models in which a pair of neural networks seek to first compress and then recreate the input data-for visualizing population genetic variation. VAEs incorporate nonlinear relationships, allow users to define the dimensionality of the latent space, and in our tests preserve global geometry better than t-SNE and UMAP. Our implementation, which we call popvae, is available as a command-line python program at ***/kr-colab/popvae. The approach yields latent embeddings that capture subtle aspects of population structure in humans and Anopheles mosquitoes, and can generate artificial genotypes characteristic of a given sample or population.

关键词： population structure population genetics data visualization pca variational autoencoder deep learning machine learning neural network

来源：评论

学校读者我要写书评

暂无评论

Open-set lung sound recognition model based on conditional Gaussian capsule network and variational time-frequency feature reconstruction

引用

BIOMEDICAL SIGNAL PROCESSING AND CONTROL 2024年第PartB期87卷

作者： Zhang, Yixuan Zhang, Jingye Shi, Lukui Hebei Univ Technol Sch Artificial Intelligence Tianjin 300401 Peoples R China

Lung sound auscultation is an essential method for diagnosing lung diseases;however, most existing lung sound recognition methods fail to identify classes that are unknown in training. Thus, we proposed an open-set lung sound recognition model based on the conditional Gaussian capsule network and variational time- frequency feature reconstruction. The proposed model incorporates an inference network, a cubic encoder, an attention module, a classifier, a cubic decoder, and a generative network. First, the inference network is employed to extract the time-frequency features of lung sounds at a single time step. Then, the variational distribution of lung sounds is computed using the capsule network and optimized to approximate the Gaussian model of the class to which the sample belongs according to the labels. Time-frequency synchronized feature extraction and reconstruction are performed on the entire lung sound sample using the cubic encoder and cubic decoder. Finally, we utilize the generative network to refactor the lung sound features for open-set recognition. The proposed model was evaluated experimentally on a combined dataset using two different category assignment schemes. The results demonstrate that the proposed model achieved accuracies of 82.31% and 88.47%, respectively, thereby outperforming existing methods.

关键词： Lung sound recognition Open-set recognition variational autoencoder Feature reconstruction

来源：评论

学校读者我要写书评

暂无评论

Generative design of graded metamaterial arrays for dynamic response modulation

引用

MATERIALS & DESIGN 2024年 237卷

作者： Wang, Weidi Cheney, Willoughby Amirkhizi, Alireza, V Univ Massachusetts Dept Mech Engn 1 Univ Ave Lowell MA 01854 USA

Mechanical metamaterials (MMs) are micro-structured systems that have long attracted research as well as application interests due to their exotic dynamic functionalities and properties not seen in ordinary materials. Graded MM arrays further enrich the scope of achievable dynamic behavior by employing non-periodic media. Designing such systems is particularly challenging due to the unclear design-performance relationship as well as the heavy computational burden. In this work, we aim to control the time domain response patterns of finite MM arrays and introduce a data-driven design approach that addresses the current challenges. A high-fidelity reduced order modeling (ROM) method is incorporated with statistical learning approaches to realize data generation and physics validation with minimal effort. A variational autoencoder (VAE) is trained to learn the design-performance relation and is used to retrieve classes of design configurations associated with a desired performance. An example application for impact mitigation is shown, and the designed MM arrays exhibit superior protection performance. The combined ROM-VAE approach presents a systematic toolset for designing graded MM arrays with modulated responses, capable of a broad spectrum of tasks such as fast prototyping, inverse generation, and design principle identification.

关键词： Metamaterial Impact mitigation variational autoencoder Reduced order model

来源：评论

学校读者我要写书评

暂无评论

CFNet: An infrared and visible image compression fusion network

引用

PATTERN RECOGNITION 2024年 156卷

作者： Xing, Mengliang Liu, Gang Tang, Haojie Qian, Yao Zhang, Jun Shanghai Univ Elect Power Sch Automat Engn Shanghai 200090 Peoples R China Shanghai JA SOLAR Technol Co Ltd Shanghai 200436 Peoples R China

Image fusion aims to acquire a more complete image representation within a limited physical space to more effectively support practical vision applications. Although the currently popular infrared and visible image fusion algorithms take practical applications into consideration. However, they did not fully consider the redundancy and transmission efficiency of image data. To address this limitation, this paper proposes a compression fusion network for infrared and visible images based on joint CNN and Transformer, termed CFNet. First of all, the idea of variational autoencoder image compression is introduced into the image fusion framework, achieving data compression while maintaining image fusion quality and reducing redundancy. Moreover, a joint CNN and Transformer network structure is proposed, which comprehensively considers the local information extracted by CNN and the global long-distance dependencies emphasized by Transformer. Finally, multi-channel loss based on region of interest is used to guide network training. Not only can color visible and infrared images be fused directly but more bits can be allocated to the foreground region of interest, resulting in a superior compression ratio. Extensive qualitative and quantitative analyses affirm that the proposed compression fusion algorithm achieves state-of-the-art performance. In particular, rate-distortion performance experiments demonstrate the great advantages of the proposed algorithm for data storage and transmission. The source code is available at https://***/Xiaoxing0503/CFNet.

关键词： Image fusion Image compression variational autoencoder Transformer Region of interest

来源：评论

学校读者我要写书评

暂无评论

A generative design framework for passive thermal control with macroscopic metamaterials

引用

THERMAL SCIENCE AND ENGINEERING PROGRESS 2024年 51卷

作者： Ignuta-Ciuncanu, Matei C. Tabor, Philip Martinez-Botas, Ricardo F. Imperial Coll London Dept Mech Engn Sustainable Energy Technol & Turbomachinery Lab London SW7 2AZ England

Efficient thermal management holds pivotal significance in the rapidly evolving realm of energy -dense applications, notably in high-performance computing, where the push toward miniaturization encounters limitations in controlling heat flux within confined spaces. The advent of metamaterials presents an energyfree opportunity to transform thermal control, essential to boosting system efficiency. This work marks a step toward integrating AI in the design and optimization of compact systems, such as printed circuit boards, solar collectors, and power electronics. We present an automated generative design exploration framework for thermal metamaterials capable of effectively controlling heat transfer to achieve multiple objectives: unperturbed temperature distribution (cloaking) and targeted heat flux insulation and concentration. The architected structures developed leverage the optimization of unit cell arrangements with contrasting thermal properties to exhibit characteristics not commonly occurring in nature. This framework employs a variational autoencoder (VAE), trained on bio-inspired leaf and mushroom microstructures, to assemble bi-material unit cells on a flexible lattice for arbitrary domains. By operating a genetic optimizer within the reduced design space of the VAE, we develop these assemblies to address adversarial temperature and heat flux objectives using a finite element solver. We demonstrate the modularity and scalability of the framework, outputting multi -objective metamaterial solutions for user -specified thermal requirements within a total run-time of 5 min only.

关键词： variational autoencoder Genetic algorithm Multi-objective optimization Heat conduction Thermal cloaking Passive thermal management

来源：评论

学校读者我要写书评

暂无评论

Nonlinear blind source separation exploiting spatial nonstationarity

引用

INFORMATION SCIENCES 2024年 665卷

作者： Sipila, Mika Nordhausen, Klaus Taskinen, Sara Univ Jyvaskyla Dept Math & Stat Jyvaskyla Finland

In spatial blind source separation the observed multivariate random fields are assumed to be mixtures of latent spatially dependent random fields. The objective is to recover latent random fields by estimating the unmixing transformation. Currently, the algorithms for spatial blind source separation can only estimate linear unmixing transformations. Nonlinear blind source separation methods for spatial data are scarce. In this paper, we extend an identifiable variational autoencoder that can estimate nonlinear unmixing transformations to spatially dependent data, and demonstrate its performance for both stationary and nonstationary spatial data using simulations. In addition, we introduce scaled mean absolute Shapley additive explanations for interpreting the latent components through nonlinear mixing transformation. The spatial identifiable variational autoencoder is applied to a geochemical dataset to find the latent random fields, which are then interpreted by using the scaled mean absolute Shapley additive explanations. Finally, we illustrate how the proposed method can be used as a pre-processing method when making multivariate predictions.

关键词： Independent component analysis Multivariate spatial data Shapley values variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Multi-rate modulation encoding via unsupervised learning for audio event detection

引用

EURASIP JOURNAL ON AUDIO SPEECH AND MUSIC PROCESSING 2024年第1期2024卷 19页

作者： Kothinti, Sandeep Reddy Elhilali, Mounya Johns Hopkins Univ Lab Computat Auditory Percept Baltimore MD 21218 USA

Technologies in healthcare, smart homes, security, ecology, and entertainment all deploy audio event detection (AED) in order to detect sound events in an audio recording. Effective AED techniques rely heavily on supervised or semi-supervised models to capture the wide range of dynamics spanned by sound events in order to achieve temporally precise boundaries and accurate event classification. These methods require extensive collections of labeled or weakly labeled in-domain data, which is costly and labor-intensive. Importantly, these approaches do not fully leverage the inherent variability and range of dynamics across sound events, aspects that can be effectively identified through unsupervised methods. The present work proposes an approach based on multi-rate autoencoders that are pretrained in an unsupervised way to leverage unlabeled audio data and ultimately learn the rich temporal dynamics inherent in natural sound events. This approach utilizes parallel autoencoders that achieve decompositions of the modulation spectrum along different bands. In addition, we introduce a rate-selective temporal contrastive loss to align the training objective with event detection metrics. Optimizing the configuration of multi-rate encoders and the temporal contrastive loss leads to notable improvements in domestic sound event detection in the context of the DCASE challenge.

关键词： Audio event detection Multi-rate processing Temporal contrastive loss Unsupervised learning variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

TopicVAE: Topic-aware Disentanglement Representation Learning for Enhanced Recommendation 22

TopicVAE: Topic-aware Disentanglement Representation Learnin...

引用

30th ACM International Conference on Multimedia (MM)

作者： Guo, Zhiqiang Li, Guohui Li, Jianjun Chen, Huaicong Huazhong Univ Sci & Technol Sch Comp Sci & Technol Wuhan Peoples R China Huazhong Univ Sci & Technol Sch Software Engn Wuhan Peoples R China

ISBN: (纸本)9781450392037

Learning disentangled representations that reflect user preference based on user behavior (implicit feedback, such as click and purchase) and content information (e.g., plot description, poster) has become a hot research topic in modern recommender systems. However, most existing methods considering content information are not well-designed to disentangle user preference features due to neglecting the diversity of user preference on different semantic topics of items, resulting in sub-optimal performance and low interpretability. To address this problem, we propose a novel Topicaware Disentangled variational autoencoder (TopicVAE) to learn disentangled representations for enhanced recommendation. Specifically, we first utilize an attention-based topic extraction to extract the topic-level item representations and topic-item probability distribution from item content, and then introduce variational autoencoder to infer topic-level disentangled user representations. To guide the learning of topic-level disentanglement, we present a topic-guided self-supervised contrastive loss to promote the otherness of different topics by introducing a neighborhood-based user representation as guidance. Besides, a heuristic regularization is designed to force each dimension of the disentangled representations to independently reflect a fine-grained factor of a specific topic (e.g., red or blue for color) for feature-level disentanglement. Extensive experimental studies on three public datasets show that TopicVAE significantly outperforms several state-of-the-art baselines. Further empirical experiments also illustrate the interpretability of disentangled representations learned by TopicVAE.

关键词： Recommendation Representation Learning Disentangled representation variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Build Connections Between Two Groups of Images Using Deep Learning Method 46

Build Connections Between Two Groups of Images Using Deep Le...

引用

46th Annual IEEE-Computer-Society International Computers, Software, and Applications Conference (COMPSAC) - Computers, Software, and Applications in an Uncertain World

作者： Du, Hongfei Barut, Emre Su, Juntao George Washington Univ Dept Stat Washington DC 20052 USA Amazon Seattle WA USA

ISBN: (纸本)9781665488105

Finding relationships between two related groups of images has been a challenging question in many fields. E.g., it is informative in neuron science studies to build the connection between experiment animals' neuron activity images and their behavior images. Very few previous works have achieved this task since most generative models focus on reconstructing the output images similar to input images. We proposed a novel framework in this paper to accomplish this goal, which could map images from one group to images from another group. We apply the singular value decomposition (SVD) method to remove the original images' background noise. Next, we combine two deep learning approaches, variational autoencoder (VAE) and convolutional neural networks (CNN), to directly connect two groups of images. We test our framework on images from a neuron science experiment. Results show that the proposed framework could generate mice paw movement images given the mice neuron images, which are very close to the ground truth images. In terms of capturing the paw gestures in paw movement images, experiment results demonstrate that our framework outperforms the state-of-art paw location detection method.

关键词： variational autoencoder Convolutional Neural Networks Singular Value Decomposition

来源：评论

学校读者我要写书评

暂无评论

SEGTRANSVAE: HYBRID CNN - TRANSFORMER WITH REGULARIZATION FOR MEDICAL IMAGE SEGMENTATION 19

SEGTRANSVAE: HYBRID CNN - TRANSFORMER WITH REGULARIZATION FO...

引用

19th IEEE International Symposium on Biomedical Imaging (IEEE ISBI)

作者： Quan-Dung Pham Hai Nguyen-Truong Nam Nguyen Phuong Nguyen, Khoa N. A. Nguyen, Chanh D. T. Bui, Trung Truong, Steven Q. H. VinBrain JSC Hanoi Vietnam Univ Sci Ho Chi Minh City Vietnam Vietnam Natl Univ Ho Chi Minh City Vietnam Vin Univ Hanoi Vietnam

ISBN: (纸本)9781665429238

Current research on deep learning for medical image segmentation exposes their limitations in learning either global semantic information or local contextual information. To tackle these issues, a novel network named SegTransVAE is proposed in this paper. SegTransVAE is built upon encoderdecoder architecture, exploiting transformer with the variational autoencoder (VAE) branch to the network to reconstruct the input images jointly with segmentation. To the best of our knowledge, this is the first method combining the success of CNN, transformer, and VAE. Evaluation on various recently introduced datasets shows that SegTransVAE outperforms previous methods in Dice Score and 95%-Haudorff Distance while having comparable inference time to a simple CNN-based architecture network. The source code is available at: https://***/itruonghai/SegTransVAE.

关键词： Transformer variational autoencoder Medical Image Segmentation MRI brain tumor CT kidney

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：