检索结果-内蒙古大学图书馆

Improving Neural Topic Models with Wasserstein Knowledge Distillation 1

45th European Conference on Information Retrieval (ECIR)

作者： Adhya, Suman Sanyal, Debarshi Kumar Indian Assoc Cultivat Sci Jadavpur 700032 India

ISBN: (数字)9783031282386

ISBN: (纸本)9783031282379;9783031282386

Topic modeling is a dominant method for exploring document collections on the web and in digital libraries. Recent approaches to topic modeling use pretrained contextualized language models and variational autoencoders. However, large neural topic models have a considerable memory footprint. In this paper, we propose a knowledge distillation framework to compress a contextualized topicmodel without loss in topic quality. In particular, the proposed distillation objective is to minimize the cross-entropy of the soft labels produced by the teacher and the student models, as well as to minimize the squared 2-Wasserstein distance between the latent distributions learned by the two models. Experiments on two publicly available datasets show that the student trained with knowledge distillation achieves topic coherence much higher than that of the original student model, and even surpasses the teacher while containing far fewer parameters than the teacher. The distilled model also outperforms several other competitive topic models on topic coherence.

关键词： Topic modeling Knowledge distillation Wasserstein distance Contextualized topic model variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Interpretable Style Transfer for Text-to-Speech with ControlVAE and Diffusion Bridge 24

Interpretable Style Transfer for Text-to-Speech with Control...

引用

Interspeech Conference

作者： Guan, Wenhao Li, Tao Li, Yishuang Huang, Hukai Hong, Qingyang Li, Lin Xiamen Univ Sch Informat Xiamen Peoples R China Xiamen Univ Inst Artificial Intelligence Xiamen Peoples R China Xiamen Univ Sch Elect Sci & Engn Xiamen Peoples R China

With the demand for autonomous control and personalized speech generation, the style control and transfer in Text-to-Speech (TTS) is becoming more and more important. In this paper, we propose a new TTS system that can perform style transfer with interpretability and high fidelity. Firstly, we design a TTS system that combines variational autoencoder (VAE) and diffusion refiner to get refined mel-spectrograms. Specifically, a two-stage and a one-stage system are designed respectively, to improve the audio quality and the performance of style transfer. Secondly, a diffusion bridge of quantized VAE is designed to efficiently learn complex discrete style representations and improve the performance of style transfer. To have a better ability of style transfer, we introduce ControlVAE to improve the reconstruction quality and have good interpretability simultaneously. Experiments on LibriTTS dataset demonstrate that our method is more effective than baseline models

关键词： speech synthesis style transfer variational autoencoder diffusion probabilistic model

来源：评论

学校读者我要写书评

暂无评论

Better Integrating Vision and Semantics for Improving Few-shot Classification 23

Better Integrating Vision and Semantics for Improving Few-sh...

引用

31st ACM International Conference on Multimedia (MM)

作者： Li, Zhuoling Wang, Yong Cent South Univ Changsha Hunan Peoples R China

ISBN: (纸本)9798400701085

Some recent methods address few-shot classification by integrating visual and semantic prototypes. However, they usually ignore the difference in feature structure between the visual and semantic modalities, which leads to limited performance improvements. In this paper, we propose a novel method, called bimodal integrator (BMI), to better integrate visual and semantic prototypes. In BMI, we first construct a latent space for each modality via a variational autoencoder, and then align the semantic latent space to the visual latent space. Through this semantics-to-vision alignment, the semantic modality is mapped to the visual latent space and has the same feature structure as the visual modality. As a result, the visual and semantic prototypes can be better integrated. In addition, based on the multivariate Gaussian distribution and the prompt engineering, a data augmentation scheme is designed to ensure the accuracy of modality alignment during the training process. Experimental results demonstrate that BMI significantly improves few-shot classification, making simple baselines outperform the most advanced methods on miniImageNet and tieredImageNet datasets.

关键词： modality alignment few-shot classification vision and semantics variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

scVAEBGM: Clustering Analysis of Single-Cell ATAC-seq Data Using a Deep Generative Model

引用

INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES 2022年第4期14卷 917-928页

作者： Duan, Hongyu Li, Feng Shang, Junliang Liu, Jinxing Li, Yan Liu, Xikui Qufu Normal Univ Sch Comp Sci Rizhao 276826 Peoples R China Shandong Univ Sci & Technol Dept Elect Engn & Informat Technol Jinan 250031 Shandong Peoples R China

A surge in research has occurred because of current developments in single-cell technologies. Above all, single-cell Assay for Transposase-Accessible Chromatin with high throughput sequencing (scATAC-seq) is a popular approach of analyzing chromatin accessibility differences at the level of single cell, either within or between groups. As a result, it is critical to examine cell heterogeneity at a previously unseen level and to identify both recognized and unknown cell types. However, with the ever-increasing number of cells engendered by technological development and the characteristics of the data, such as high noise, sparsity and dimension, challenges in distinguishing cell types have emerged. We propose scVAEBGM, which integrates a variational autoencoder (VAE) with a Bayesian Gaussian-mixture model (BGM) to process and analyze scATAC-seq data. This method combines and takes benefits of a Bayesian Gaussian mixture model to estimate the number of cell types without determining the cluster number in a beforehand. In other words, the size of the clusters is inferred from the data, thus avoiding biases introduced by subjective assessments when manually determining the size of the clusters. Additionally, the method is more robust to noise and can better represent single-cell data in lower dimensions. We also create a further clustering strategy. It is indicated by experiments that further clustering based on the already completed clustering can improve the clustering accuracy again. We test on six public datasets, and scVAEBGM outperforms various dimension reduction baselines. In downstream applications, scVAEBGM can reveal biological cell types. [GRAPHICS] .

关键词： scATAC-seq Clustering Deep learning variational autoencoder Bayesian Gaussian-mixture model

来源：评论

学校读者我要写书评

暂无评论

A privacy-preserving vehicle trajectory clustering framework

引用

Frontiers of Information Technology & Electronic Engineering 2024年第7期25卷 988-1002页

作者： Ran TIAN Pulun GAO Yanxing LIU College of Computer Science&Engineering Northwest Normal UniversityLanzhou 730070China

As one of the essential tools for spatio‒temporal traffic data mining,vehicle trajectory clustering is widely used to mine the behavior patterns of ***,uploading original vehicle trajectory data to the server and clustering carry the risk of privacy ***,one of the current challenges is determining how to perform vehicle trajectory clustering while protecting user *** propose a privacy-preserving vehicle trajectory clustering framework and construct a vehicle trajectory clustering model(IKV)based on the variational autoencoder(VAE)and an improved K-means *** the framework,the client calculates the hidden variables of the vehicle trajectory and uploads the variables to the server;the server uses the hidden variables for clustering analysis and delivers the analysis results to the *** IKV’workflow is as follows:first,we train the VAE with historical vehicle trajectory data(when VAE’s decoder can approximate the original data,the encoder is deployed to the edge computing device);second,the edge device transmits the hidden variables to the server;finally,clustering is performed using improved K-means,which prevents the leakage of the vehicle *** is compared to numerous clustering methods on three *** the nine performance comparison experiments,IKV achieves optimal or sub-optimal performance in six of the ***,in the nine sensitivity analysis experiments,IKV not only demonstrates significant stability in seven experiments but also shows good robustness to hyperparameter *** results validate that the framework proposed in this paper is not only suitable for privacy-conscious production environments,such as carpooling tasks,but also adapts to clustering tasks of different magnitudes due to the low sensitivity to the number of cluster centers.

关键词： Privacy protection variational autoencoder Improved K-means Vehicle trajectory clustering

来源：评论

学校读者我要写书评

暂无评论

Few Shot Hematopoietic Cell Classification 6

Few Shot Hematopoietic Cell Classification

引用

6th International Conference on Medical Imaging with Deep Learning (MIDL)

作者： Nguyen, Vu Howlader, Prantik Hou, Le Samaras, Dimitris Gupta, Rajarsi Saltz, Joel SUNY Stony Brook Dept Comp Sci Stony Brook NY USA SUNY Stony Brook Dept Biomed Informat Stony Brook NY USA

We propose a few shot learning approach for the problem of hematopoietic cell classification in digital pathology. In hematopoiesis cell classification, the classes correspond to the different stages of the cellular maturation process. Two consecutive stage categories are considered to have a neighborhood relationship, which implies a visual similarity between the two categories. We propose RelationVAE which incorporates these relationships between hematopoietic cell classes to robustly generate more data for the classes with limited training data. Specifically, we first model these relationships using a graphical model, and propose RelationVAE, a deep generative model which implements the graphical model. RelationVAE is trained to optimize the lower bound of the pairwise data likelihood of the graphical model. In this way, it can identify class level features of a specific class from a small number of input images together with the knowledge transferred from visually similar classes, leading to more robust sample synthesis. The experiments on our collected hematopoietic dataset show the improved results of our proposed RelationVAE over a baseline VAE model and other few shot learning methods. Our code and data are available at https://***/cvlab-stonybrook/hematopoiesis-relationvae.

关键词： Hematopoiesis cell classification Few Shot Learning variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

An efficient dimensionality reduction method using filter-based feature selection and variational autoencoders on Parkinson's disease classification

引用

BIOMEDICAL SIGNAL PROCESSING AND CONTROL 2021年第0期66卷 102452-102452页

作者： Gunduz, Hakan Bandirma Onyedi Eylul Univ Engn & Nat Sci Fac Software Engn Dept TR-10200 Bandirma Balikesir Turkey

Parkinson's disease (Pd) is a progressive disease caused by the loss of brain cells and brings about speech and pronunciation defects during the early stages. This study revealed a Pd classification system based on vocal features extracted from the voice recordings of the individuals and proposed a hybrid dimensionality reduction methods to extract robust features. Proposed method took advantage of the prominent aspects of variational autoencoders (VAE) and filter-based feature selection models. Relief and Fisher Score were selected as filter-based methods for their effective performance in handling noisy data while VAE was used as a feature extractor due to the capability of preserving the regular latent space properties during the feature generation. In order to assess the effectiveness of the devised method, multi-kernel Support Vector Machines (SVM) classifier were trained with obtained deep feature representations. The combination of deep Relief features and SVM with multiple kernels distinguished Pd individuals from healthy subjects with an accuracy of 0.916 with 0.772 Matthews Correlation Coefficient (MCC) rates using only 30 features. Compared to results obtained without dimensionality reduction, proposed model provided approximately 9% and 22% improvements on accuracy and MCC rates, respectively. All experimental results showed that models trained with the deep features had higher accuracy and MCC rates with those trained with Fisher Score and Relief selected features. In addition, all models trained with reduced features had higher classification performance than the model without selection. It was also concluded that using multiple kernels in the SVM boosted the classification performance.

关键词： Parkinson's disease prediction Dimensionality reduction variational autoencoder Fisher score Relief Multi-Kernel SVM

来源：评论

学校读者我要写书评

暂无评论

Measurement Error Models: From Nonparametric Methods to Deep Neural Networks

引用

STATISTICAL SCIENCE 2022年第4期37卷 473-493页

作者： Hu, Zhirui Ke, Zheng Tracy Liu, Jun S. Tsinghua Univ Ctr Stat Sci Beijing Peoples R China Harvard Univ Dept Stat Cambridge MA 02138 USA

The success of deep learning has inspired a lot of recent interests in exploiting neural network structures for statistical inference and learning. In this paper, we review some popular deep neural network structures and techniques under the framework of nonparametric regression with measurement errors. In particular, we demonstrate how to use a fully connected feed-forward neural network to approximate the regression function f (x), explain how to use a normalizing flow to approximate the prior distribution of X, and detail how to construct an inference network to generate approximate posterior samples of X. After reviewing recent advances in variational inference for deep neural networks, such as the importance weighted autoencoder, doubly reparametrized gradient estimator, and nonlinear independent components estimation, we describe an inference procedure built upon these advances. An extensive numerical study is presented to compare the neural network approach with classical nonparametric methods, which suggests that the neural network approach is more flexible in accommodating different classes of regression functions and performs superior or comparable to the best available method in many settings.

关键词： Doubly reparametrized gradient estimator errorin-variables fully connected feed-forward neural network importance weighted autoencoder normalizing flow SIMEX variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Adapting RICO dataset for boosting Graphical User Interface component classification for automated Android testing 10

Adapting RICO dataset for boosting Graphical User Interface ...

引用

10th International Conference on Soft Computing and Machine Intelligence (ISCMI)

作者： Lopez, Yadini Perez Albuquerque, Lais Dib de F. Costa Junior, Gilmar Joia Xavier, Daniel Lopes Ochoa, Juan David Camargo, Denizard Dimitri SIDIA Dev & Res Inst SITA AIG Automat Innovat Grp Manaus Amazonas Brazil SIDIA Dev & Res Inst SITA AIG Automat Innovat Grp Porto Velho Brazil

ISBN: (纸本)9798350359374

User Interface testing is one of the most used routines for Android feature quality verification. Regarding this, several industrial and academic solutions are available for Graphical User Interface component identification. However, the state-of-art works are still limited respecting the variety of widgets that can recognize. Also, most solutions use application metadata that can vary across different releases and device models, affecting the portability of automations. Research that approach widget classification using Machine Learning based solutions commonly use a large Android screen capture dataset called RICO. However, several annotations problems have been recurrently pointed on this dataset. In this work, we propose a selection of data cleaning and balancing techniques for removing noisy samples and leveling the number of samples per class on RICO dataset. We used a custom dataset with an extended number of classes (106) when comparing with the state-of-art works that approach widget classification with 15 classes on average. Finally, we showed that using these techniques to improve the quality of data can improve the accuracy when training Machine Learning models like Convolutional Neural Networks and eXtreme Gradient Boosting.

关键词： Android automated mobile testing variational autoencoder widget graphical user interface dataset

来源：评论

学校读者我要写书评

暂无评论

Interpretable Mechanistic Representations for Meal-level Glycemic Control in the Wild 3

Interpretable Mechanistic Representations for Meal-level Gly...

引用

3rd Machine Learning for Health Symposium

作者： Wang, Ke Alexander Fox, Emily B. Stanford Univ Stanford CA 94305 USA CZ Biohub San Francisco CA USA

Diabetes encompasses a complex landscape of glycemic control that varies widely among individuals. However, current methods do not faithfully capture this variability at the meal level. On the one hand, expert-crafted features lack the flexibility of data-driven methods;on the other hand, learned representations tend to be uninterpretable which hampers clinical adoption. In this paper, we propose a hybrid variational autoencoder to learn interpretable representations of CGM and meal data. Our method grounds the latent space to the inputs of a mechanistic differential equation, producing embeddings that reflect physiological quantities, such as insulin sensitivity, glucose effectiveness, and basal glucose levels. Moreover, we introduce a novel method to infer the glucose appearance rate, making the mechanistic model robust to unreliable meal logs. On a dataset of CGM and self-reported meals from individuals with type-2 diabetes and pre-diabetes, our unsupervised representation discovers a separation between individuals proportional to their disease severity. Our embeddings produce clusters that are up to 4x better than naive, expert, black-box, and pure mechanistic features. Our method provides a nuanced, yet interpretable, embedding space to compare glycemic control within and across individuals, directly learnable from in-the-wild data.

关键词： interpretable representations mechanistic differential equations CGM diabetes T2D glycemic variability disease subtyping disease progression variational autoencoder hybrid modeling

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：