检索结果-内蒙古大学图书馆

34th Great Lakes Symposium on VLSI (GLSVLSI)

作者： Liao, Yuchao Adegbija, Tosiron Lysecky, Roman Tandon, Ravi Univ Arizona Elect & Comp Engn Tucson AZ 85721 USA

ISBN: (纸本)9798400706059

High-Level Synthesis (HLS) Design Space Exploration (DSE) is a widely accepted approach for efficiently exploring Pareto-optimal and optimal hardware solutions during the HLS process. Several HLS benchmarks and datasets are available for the research community to evaluate their methodologies. Unfortunately, these resources are limited and may not be sufficient for complex, multi-component system-level explorations. Generating new data using existing HLS benchmarks can be cumbersome, given the expertise and time required to effectively generate data for different HLS designs and directives. As a result, synthetic data has been used in prior work to evaluate system-level HLS DSE. However, the fidelity of the synthetic data to real data is often unclear, leading to uncertainty about the quality of system-level HLS DSE. This paper proposes a novel approach, called Vaegan, that employs generative machine learning to generate synthetic data that is robust enough to support complex system-level HLS DSE experiments that would be unattainable with only the currently available data. We explore and adapt a variational autoencoder (VAE) and Generative Adversarial Network (GAN) for this task and evaluate our approach using state-of-the-art datasets and metrics. We compare our approach to prior works and show that Vaegan effectively generates synthetic HLS data that closely mirrors the ground truth's distribution.

关键词： High-Level Synthesis Synthetic Data Generation variational autoencoder Generative Adversarial Network

来源：评论

学校读者我要写书评

暂无评论

VITS: Quality Vs. Speed Analysis 26th

VITS: Quality Vs. Speed Analysis

引用

26th Annual International Conference on Text, Speech and Dialogue (TSD)

作者： Matousek, Jindrich Tihelka, Daniel Univ West Bohemia Fac Appl Sci Dept Cybernet Plzen Czech Republic Univ West Bohemia Fac Appl Sci New Technol Informat Soc NTIS Plzen Czech Republic

ISBN: (纸本)9783031404979;9783031404986

In this paper, we analyze the performance of a modern end-to-end speech synthesis model called variational Inference with adversarial learning for end-to-end Text-to-Speech (VITS). We build on the original VITS model and examine how different modifications to its architecture affect synthetic speech quality and computational complexity. Experiments with two Czech voices, a male and a female, were carried out. To assess the quality of speech synthesized by the different modified models, MUSHRA listening tests were performed. The computational complexity was measured in terms of synthesis speed over real time. While the original VITS model is still preferred regarding speech quality, we present a modification of the original structure with a significantly better response yet providing acceptable output quality. Such a configuration can be used when system response latency is critical.

关键词： Neural speech synthesis End-to-end modeling variational autoencoder VITS Speed optimization

来源：评论

学校读者我要写书评

暂无评论

IMPORTANCE SAMPLING BASED FEDERATED UNSUPERVISED REPRESENTATION LEARNING 49

IMPORTANCE SAMPLING BASED FEDERATED UNSUPERVISED REPRESENTAT...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Shaht, Nazreen Goyalt, Prachi Prasad, Ranjitha IIITD ECE Dept Delhi India

ISBN: (纸本)9798350344868;9798350344851

The use of AI has led to the era of pervasive intelligence, marked by a proliferation of smart devices in our daily lives. Federated Learning (FL) enables machine learning at the edge without having to share user-specific private data with an untrusted third party. Conventional FL techniques are supervised learning methods, where a fundamental challenge is to ensure that data is reliably annotated at the edge. Another approach is to obtain rich and informative representations of unlabeled data, which is suitable for downstream tasks. We propose a novel IS-FedVAE framework where we use importance sampling to federate a global VAE framework, allowing us to learn the global latent space distribution using local latent space distributions at the edge. We evaluate the representation in a stand-alone manner using linear probe, where we freeze the backbone representation and measure the accuracy of a downstream classifier. Furthermore, we demonstrate that IS-FedVAE outperforms state-of-the-art unsupervised FL learning baselines. We also show that IS-FedVAE is insensitive to varying levels of heterogeneity, scalable to varying numbers of clients, and robust to the changing number of local epochs.

关键词： variational autoencoder representation learning federated learning importance sampling.

来源：评论

学校读者我要写书评

暂无评论

A STUDY ON THE EFFECT OF COLOR SPACES IN LEARNED IMAGE COMPRESSION 31

A STUDY ON THE EFFECT OF COLOR SPACES IN LEARNED IMAGE COMPR...

引用

2024 International Conference on Image Processing

作者： Prativadibhayankaram, Srivatsa Panda, Mahadev Prasad Seiler, Juergen Richter, Thomas Sparenberg, Heiko Foessel, Siegfried Kaup, Andre Fraunhofer Inst Integrated Circuits IIS Moving Picture Technol Erlangen Germany Friedrich Alexander Univ Erlangen Nurnberg Multimedia Commun & Signal Proc Erlangen Germany RheinMain Univ Appl Sci Media Concept & Prod Wiesbaden Germany

ISBN: (纸本)9798350349405;9798350349399

In this work, we present a comparison between color spaces namely YUV, LAB, RGB and their effect on learned image compression. For this we use the structure and color based learned image codec (SLIC) from our prior work, which consists of two branches - one for the luminance component (Y or L) and another for chrominance components (UV or AB). However, for the RGB variant we input all 3 channels in a single branch, similar to most learned image codecs operating in RGB. The models are trained for multiple bitrate configurations in each color space. We report the findings from our experiments by evaluating them on various datasets and compare the results to state-of-the-art image codecs. The YUV model performs better than the LAB variant in terms of MS-SSIM with a Bjontegaard delta bitrate (BD-BR) gain of 7.5% using VTM intra-coding mode as the baseline. Whereas the LAB variant has a better performance than YUV model in terms of CIEDE2000 having a BD-BR gain of 8%. Overall, the RGB variant of SLIC achieves the best performance with a BD-BR gain of 13.14% in terms of MS-SSIM and a gain of 17.96% in CIEDE2000 at the cost of a higher model complexity.

关键词： Deep learning learned image compression image compression variational autoencoder color learning color spaces

来源：评论

学校读者我要写书评

暂无评论

Physics-based Character Controllers Using Conditional VAEs

引用

ACM TRANSACTIONS ON GRAPHICS 2022年第4期41卷 p1-12页

作者： Won, Jungdam Gopinath, Deepak Hodgins, Jessica Meta AI Pittsburgh PA 15211 USA Carnegie Mellon Univ Pittsburgh PA 15213 USA

High-quality motion capture datasets are now publicly available, and researchers have used them to create kinematics-based controllers that can generate plausible and diverse human motions without conditioning on specific goals (i.e., a task-agnostic generative model). In this paper, we present an algorithm to build such controllers for physically simulated characters having many degrees of freedom. Our physics-based controllers are learned by using conditional VAEs, which can perform a variety of behaviors that are similar to motions in the training dataset. The controllers are robust enough to generate more than a few minutes of motion without conditioning on specific goals and to allow many complex downstream tasks to be solved efficiently. To show the effectiveness of our method, we demonstrate controllers learned from several different motion capture databases and use them to solve a number of downstream tasks that are challenging to learn controllers that generate natural-looking motions from scratch. We also perform ablation studies to demonstrate the importance of the elements of the algorithm.

关键词： Character Animation Physics-based Simulation and Control Motion Capture Reinforcement Learning variational autoencoder Behavior Cloning

来源：评论

学校读者我要写书评

暂无评论

Classification of Motor Imagery EEG Signals Based on Deep autoencoder and Convolutional Neural Network Approach

引用

IEEE ACCESS 2022年 10卷 48071-48081页

作者： Hwaidi, Jamal F. Chen, Thomas M. Univ London Dept Elect & Elect Engn London EC1V 0HB England

The technology of the brain-computer interface (BCI) employs electroencephalogram (EEG) signals to establish direct interaction between the human body and its surroundings with promising applications in medical rehabilitative services and cognitive science. Deep learning approaches, particularly the detection and analysis of motor imagery signals using convolutional neural network (CNN) frameworks have produced outstanding results in the BCI system in recent years. The complex process of data representation, on the other hand, limits practical applications, and the end-to-end approach reduces the accuracy of recognition. Moreover, since noise and other signal sources can interfere with brain electrical capacitance, EEG classifiers are difficult to improve and have limited generalisation ability. To address these issues, this paper proposes a new approach for EEG motor imagery signal classification by using a variational autoencoder to remove noise from the signals, followed by a combination of deep autoencoder (DAE) and a CNN architecture to classify EEG motor imagery signals which is capable of training a deep neural network to replicate its input to output using encoding and decoding operations. Experimental results show that the proposed approach for motor imagery EEG signal classification is feasible and that it outperforms current CNN-based approaches and several traditional machine learning approaches.

关键词： Electroencephalography Feature extraction Convolutional neural networks Deep learning Support vector machines Training Brain modeling Electroencephalography deep autoencoder convolutional neural network variational autoencoder motor imagery

来源：评论

学校读者我要写书评

暂无评论

VAE-based Phoneme Alignment Using Gradient Annealing and SSL Acoustic Features 25

VAE-based Phoneme Alignment Using Gradient Annealing and SSL...

引用

25th Interspeech Conference

作者： Koriyama, Tomoki CyberAgent Tokyo Japan

This paper presents an accurate phoneme alignment model that aims for speech analysis and video content creation. We propose a variational autoencoder (VAE)-based alignment model in which a probable path is searched using encoded acoustic and linguistic embeddings in an unsupervised manner. Our proposed model is based on one TTS alignment (OTA) and extended to obtain phoneme boundaries. Specifically, we incorporate a VAE architecture to maintain consistency between the embedding and input, apply gradient annealing to avoid local optimum during training, and introduce a self-supervised learning (SSL)-based acoustic-feature input and state-level linguistic unit to utilize rich and detailed information. Experimental results show that the proposed model generated phoneme boundaries closer to annotated ones compared with the conventional OTA model, the CTC-based segmentation model, and the widely-used tool MFA.

关键词： phoneme alignment annealing self-supervised learning forward-sum algorithm variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection in Stock Market Transactions: A Comparison of Deep Learning Methods 32

Anomaly Detection in Stock Market Transactions: A Comparison...

引用

32nd IEEE Signal Processing and Communications Applications Conference (SIU)

作者： Oner, Sultan Ceren Sahan, Huseyin Demirdag, Melike Bayrak, Ahmet Tugrul Ata Teknol Platformlari Veri Bilimi & Inovasyon Istanbul Turkiye

ISBN: (纸本)9798350388978;9798350388961

This study applies unsupervised machine learning techniques and deep learning models to evaluate stock market users' past transactions to detect anomalies in transaction data regarding volume and count. First, traditional unsupervised anomaly detection methods such as support vector machines (SVM) and local outlier factor (LOF) are utilized to identify anomalies in transaction data by evaluating only past transactions and determining whether they are abnormal. This labeling enabled deep learning methods to validate and predict these transactions. Additionally, deep learning methods such as LSTM-based autoencoders, isolation forest-based LSTM, generative adversarial networks (GAN), and variational autoencoders (VAE) are employed to analyze users' past transactions and predict the number and magnitude of future transactions. This comprehensive approach combines deep learning models with traditional methods to provide infrastructure for retrospective evaluation of transaction data anomalies and prediction of future user actions.

关键词： anomaly detection long short-term memory autoencoders generative adversarial networks variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Trace-SPC anomaly detection solution based on VAE model 35

Unsupervised Trace-SPC anomaly detection solution based on V...

引用

35th Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC)

作者： Wang, Xiaofei Jiang, Hui Semitronix Corp Hangzhou Peoples R China

ISBN: (纸本)9798350384567;9798350384550

Wafer manufacturing is a complex, expensive, and time-consuming process that involves multiple steps. By closely monitoring the corresponding variations in process parameters, potential anomalies and faults in the process can be timely detected and identified, effectively improving equipment utilization and product yield. However, the anomaly detection methods currently in use rely on extracting feature values from sensor data and analyzing the statistical information of these features. This approach fails to identify complex anomaly patterns, and constructing meaningful feature values can be challenging. In this paper, we propose an unsupervised anomaly detection solution based on the variational autoencoder (VAE) model. We optimized the training strategy of the VAE model to address the issue of posterior collapse. By reconstructing the sensor data using the VAE model, we can effectively convert anomalous patterns into normal ones. The experimental results confirm the effectiveness of the model across various sensor datasets, demonstrating the capability of the proposed solution to accurately identify abnormal patterns within the sensor data.

关键词： wafer manufacturing data reconstruction Trace-SPC anomaly detection variational autoencoder

来源：评论

学校读者我要写书评

暂无评论

Four-Legged Gait Control via the Fusion of Computer Vision and Reinforcement Learning 27

Four-Legged Gait Control via the Fusion of Computer Vision a...

引用

27th International Conference on Information Fusion (FUSION)

作者： Dassori, Ignacio Adams, Martin Vasquez, Jorge Univ Chile Elect Engn Dept Santiago Chile Carnegie Mellon Univ Mech Engn Dept Pittsburgh PA 15213 USA

ISBN: (纸本)9798350371420;9781737749769

This article explores the integration of fully autonomous legged robots in obstacle filled environments, simultaneously addressing the challenges of navigation and control. Despite the potential of legged robots for dynamic tasks, their deployment in complex environments has been hindered by the difficulty of developing effective autonomous control systems. In particular, the motion planning problem is addressed in this article, by formulating it as a Partially Observable Markov Decision Process (POMDP) and applying Proximal Policy Optimization (PPO), a model-free Deep Reinforcement Learning (DRL) algorithm. To improve sample efficiency and real-world applicability, the proposed method incorporates a Central Pattern Generator (CPG) for motion planning and a variational autoencoder (VAE) for terrain representation, reducing the complexity of action and observation spaces. Referred to as the VAE-CPG architecture, its performance is demonstrated using the Unitree Laikago robot within the PyBullet simulation environment, aiming to show its effectiveness in simulated construction sites. Our findings indicate that by reducing the legged action space to periodic gait patterns and optimizing the gait based on sensory feedback, we achieve enhanced adaptability and efficiency. This work presents a viable means towards the deployment of autonomous legged robots and their improved efficiency in real applications.

关键词： Robotics Reinforcement Learning Computer Vision Central Pattern Generator variational autoencoder Gait optimization

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：