检索结果-内蒙古大学图书馆

Mix-GEMM: Extending RISC-V CPUs for Energy-Efficient Mixed-Precision DNN Inference Using binary segmentation

IEEE TRANSACTIONS ON COMPUTERS 2025年第2期74卷 582-596页

作者： Fornt, Jordi Reggiani, Enrico Fontova-Muste, Pau Rodas, Narcis Pappalardo, Alessandro Unsal, Osman Sabri Kestelman, Adrian Cristal Altet, Josep Moll, Francesc Abella, Jaume Barcelona Supercomp Ctr BSC Barcelona 08034 Spain Univ Politecn Catalunya UPC Barcelona 08034 Spain Adv Microdevices AMD Dublin D24 T683 Ireland

Efficiently computing Deep Neural Networks (DNNs) has become a primary challenge in today's computers, especially on devices targeting mobile or edge applications. Recent progress on Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT) has shown that the key to high energy efficiency lies in executing deep learning models with low- (8- to 5-bit) or ultra-low-precision (4- to 2-bit). Unfortunately, current Central Processing Unit (CPU) architectures and Instruction Set Architectures (ISAs) present severe limitations on the range of data sizes supported to compute DNN kernels. In this work, we present Mix-GEMM, a hardware-software co-designed architecture that enables RISC-V processors to efficiently compute arbitrary mixed-precision DNN kernels, supporting all data size combinations from 8- to 2-bit. By applying binary segmentation, our architecture can scale its throughput by decreasing the data size of the operands, resulting in a flexible approach capable of leveraging state-of-the-art QAT and PTQ to achieve high energy efficiency at a very low cost. Evaluating our Mix-GEMM architecture in a dual-issue in-order RISC-V processor shows that we are able to boost its performance and energy efficiency by up to 44x and 11x with respect to the baseline processor, with an area overhead of only 2%. This allows our extended processor to execute state-of-the-art DNNs with significantly higher performance and energy efficiency than the standard FP32 precision, while retaining almost the same model accuracy.

关键词： Program processors Computer architecture Energy efficiency Quantization (signal) Artificial neural networks Computers Computational modeling Pipelines Software Single instruction multiple data Deep neural networks RISC-V extensions energy efficiency binary segmentation neural accelerators

来源：评论

学校读者我要写书评

暂无评论

Advancing underwater binary image segmentation with PDBU-Net: A progressive approach to feature extraction and accuracy improvement

引用

REGIONAL STUDIES IN MARINE SCIENCE 2025年 83卷

作者： George, Geomol Anusuya, S. SIMATS Engn Chennai India SIMATS Deemed Univ Chennai Tamil Nadu India

Precise binary image segmentation in underwater environments is crucial for marine biology and underwater robotics applications, as fluctuations in lighting, water conditions, and object appearance provide substantial difficulties. To tackle these problems, this research presents PDBU-Net, an innovative deep-learning framework specifically created to improve the extraction of features and the accuracy of segmentation for binary images captured underwater. PDBU-Net incorporates a progressive feature extraction methodology into a strong deep neural network structure, resulting insignificant enhancements in segmentation performance compared to current methods. PDBU-Net was tested extensively and achieved an average Intersection over Union (IoU) of 90.39% and an F-score of 94.94%. The model achieved precision and recall rates of 94.55% and 95.39% respectively, along with an overall accuracy of 95.66%. The results validate the efficacy of PDBU-Net in precisely detecting objects in a wide range of underwater photos, showcasing its successful implementation in real underwater analysis and robotics.

关键词： Underwater images binary segmentation Progressive DenseNet Deep learning Encoder Decoder

来源：评论

学校读者我要写书评

暂无评论

Novel electromyography signal envelopes based on binary segmentation

引用

BIOMEDICAL SIGNAL PROCESSING AND CONTROL 2018年 45卷 225-236页

作者： Guerrero, J. A. Castillo-Galvan, M. A. Macias-Diaz, J. E. Univ Autonoma Aguascalientes Dept Estadist Ave Univ 940Ciudad Univ Aguascalientes 20131 Ags Mexico Univ Autonoma Aguascalientes Ctr Ciencias Basicas Ave Univ 940Ciudad Univ Aguascalientes 20131 Ags Mexico Univ Autonoma Aguascalientes Dept Matemat & Fis Ave Univ 940Ciudad Univ Aguascalientes 20131 Ags Mexico

In this work, we introduce two novel methodologies to compute the envelope of superficial electromyography signals. Our methods are based on the detection of activation and deactivation patterns using a change-point approach on the variances of the sample. More concretely, an iterative algorithms is proposed to select the change-points between two segments of the signal based on some local statistics introduced in this work. The signal is split up into two segments, and a new search for change-points is recursively conducted in each subsequence. The change -points make possible to calculate local envelopes which reflect the shape of the signal without ignoring the activation and deactivation landmarks. Two methods are proposed in this work, and the improvements with respect to methodologies available in the literature are shown using both synthetic and real data. A thorough analysis of the techniques is performed to that end. (C) 2018 Elsevier Ltd. All rights reserved.

关键词： Electromyography signal Envelope binary segmentation Change-points

来源：评论

学校读者我要写书评

暂无评论

Towards Diverse binary segmentation via a Simple yet General Gated Network

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2024年第10期132卷 4157-4234页

作者： Zhao, Xiaoqi Pang, Youwei Zhang, Lihe Lu, Huchuan Zhang, Lei Dalian Univ Technol Dalian Peoples R China Hong Kong Polytech Univ Dept Comp Kowloon Hong Kong Peoples R China OPPO Res Chengdu Peoples R China

In many binary segmentation tasks, most CNNs-based methods use a U-shape encoder-decoder network as their basic structure. They ignore two key problems when the encoder exchanges information with the decoder: one is the lack of interference control mechanism between them, the other is without considering the disparity of the contributions from different encoder levels. In this work, we propose a simple yet general gated network (GateNet) to tackle them all at once. With the help of multi-level gate units, the valuable context information from the encoder can be selectively transmitted to the decoder. In addition, we design a gated dual branch structure to build the cooperation among the features of different levels and improve the discrimination ability of the network. Furthermore, we introduce a "Fold" operation to improve the atrous convolution and form a novel folded atrous convolution, which can be flexibly embedded in ASPP or DenseASPP to accurately localize foreground objects of various scales. GateNet can be easily generalized to many binary segmentation tasks, including general and specific object segmentation and multi-modal segmentation. Without bells and whistles, our network consistently performs favorably against the state-of-the-art methods under 10 metrics on 33 datasets of 10 binary segmentation tasks.

关键词： binary segmentation Gated network Gated dual branch Folded atrous convolution

来源：评论

学校读者我要写书评

暂无评论

Functional Pattern-Related Anomaly Detection Approach Collaborating binary segmentation with Finite State Machine

引用

Computers, Materials & Continua 2023年第12期77卷 3573-3592页

作者： Ming Wan Minglei Hao Jiawei Li Jiangyuan Yao Yan Song School of Information Liaoning UniversityShenyang110036China School of Computer Science and Technology Hainan UniversityHaikou570228China School of Physics Liaoning UniversityShenyang110036China

The process control-oriented threat,which can exploit OT(Operational Technology)vulnerabilities to forcibly insert abnormal control commands or status information,has become one of the most devastating cyber attacks in industrial automation *** effectively detect this threat,this paper proposes one functional pattern-related anomaly detection approach,which skillfully collaborates the BinSeg(binary segmentation)algorithm with FSM(Finite State Machine)to identify anomalies between measuring data and control *** detecting the change points of measuring data,the BinSeg algorithm is introduced to generate some initial sequence segments,which can be further classified and merged into different functional patterns due to their backward difference means and *** analyzing the pattern association according to the Bayesian network,one functional state transition model based on FSM,which accurately describes the whole control and monitoring process,is constructed as one feasible detection ***,we use the typical SWaT(Secure Water Treatment)dataset to evaluate the proposed approach,and the experimental results show that:for one thing,compared with other change-point detection approaches,the BinSeg algorithm can be more suitable for the optimal sequence segmentation of measuring data due to its highest detection accuracy and least consuming time;for another,the proposed approach exhibits relatively excellent detection ability,because the average detection precision,recall rate and F1-score to identify 10 different attacks can reach 0.872,0.982 and 0.896,respectively.

关键词： Process control-oriented threat anomaly detection binary segmentation FSM

来源：评论

学校读者我要写书评

暂无评论

Algorithms for finding attribute value group for binary segmentation of categorical databases

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2002年第6期14卷 1269-1279页

作者： Morimoto, Y Fukuda, T Tokuyama, T IBM Corp Tokyo Res Lab Kanagawa 2428502 Japan GSIS Tohoku Univ Aoba Ku Sendai Miyagi 9808579 Japan

We consider the problem of finding a set of attribute values that give a high quality binary segmentation of a database. The quality of a segmentation is defined by an objective function suitable for the user's objective, such as "mean squared error," "mutual information," or "chi(2)," each of which is defined in terms of the distribution of a given target attribute. Our goal is to find value groups on a given conditional domain that split databases into two segments, optimizing the value of an objective function. Though the problem is intractable for general objective functions, there are feasible algorithms for finding high quality binary segmentations when the objective function is convex, and we prove that the typical criteria mentioned above are all convex. We propose two practical algorithms, based on computational geometry techniques, which find a much better value group than conventional heuristics.

关键词： value groups binary segmentation categorical test decision tree data reduction data mining

来源：评论

学校读者我要写书评

暂无评论

Pixel U-Net: an improved version of U-Net for binary segmentation of wind turbine blades

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2024年第8-9期18卷 6299-6307页

作者： Rizvi, Syed Zeeshan Jamil, Mohsin Huang, Weimin Mem Univ Newfoundland Dept Elect & Comp Engn 230 Elizabeth Ave St John NF A1C 5S7 Canada

Wind turbine technicians face significant risks of fatal injuries, which can be mitigated by utilizing drones to capture images of wind turbine blades (WTBs) for remote inspection and maintenance. Different computer vision and image processing techniques can be applied to these captured drone images to automate the process of WTB inspection. However, the captured drone images consist of challenging backgrounds, due to which the WTB area needs to be extracted from the image as a pre-processing step. This paper introduces Pixel U-Net, an enhanced U-Net architecture tailored for binary segmentation of WTB images, improving segmentation accuracy with pixel shuffle and unshuffle operations. Evaluated against baseline U-Net architecture and its variations using the publicly available Blade30 dataset, Pixel U-Net achieves an average validation accuracy of 99.0% and surpasses existing methods in WTB image segmentation. Additionally, novel architectural variations, namely U-Net + HS-Block + Pixel Unshuffle and Pixel U-Net + Attention, have also been proposed in this study, which exhibit superior performance with an average training loss of 0.012 and an average testing accuracy of 98.3%, respectively. Qualitative comparisons of the results further highlight the efficacy of deep learning-based segmentation techniques in advancing wind turbine inspection and maintenance practices.

关键词： binary segmentation Semantic segmentation Renewable energy Wind turbines Automatic inspection

来源：评论

学校读者我要写书评

暂无评论

Ensemble binary segmentation for irregularly spaced data with change-points

引用

JOURNAL OF THE KOREAN STATISTICAL SOCIETY 2022年第1期51卷 65-86页

作者： Korkas, Karolos K. Univ Greenwich Sch Comp & Math Sci London England

We propose a new technique for consistent estimation of the number and locations of the change-points in the structure of an irregularly spaced time series. The core of the segmentation procedure is the ensemble binary segmentation method (EBS), a technique in which a large number of multiple change-point detection tasks using the binary segmentation method are applied on sub-samples of the data of differing lengths, and then the results are combined to create an overall answer. We do not restrict the total number of change-points a time series can have, therefore, our proposed method works well when the spacings between change-points are short. Our main change-point detection statistic is the time-varying autoregressive conditional duration model on which we apply a transformation process in order to decorrelate it. To examine the performance of EBS we provide a simulation study for various types of scenarios. A proof of consistency is also provided. Our methodology is implemented in the R package eNchange, available to download from CRAN.

关键词： Multiple change-point detection Conditional duration models binary segmentation Ensemble methods Non-stationarity

来源：评论

学校读者我要写书评

暂无评论

Seeded binary segmentation: a general methodology for fast and optimal changepoint detection

引用

BIOMETRIKA 2023年第1期110卷 249-256页

作者： Kovacs, S. Buehlmann, P. Li, H. Munk, A. Swiss Fed Inst Technol Dept Math Seminar Stat Ramistr 101 CH-8092 Zurich Switzerland Univ Gottingen Inst Math Stochast Goldschmidtstr 7 D-37077 Gottingen Germany

We propose seeded binary segmentation for large-scale changepoint detection problems. We construct a deterministic set of background intervals, called seeded intervals, in which single changepoint candidates are searched for. The final selection of changepoints based on these candidates can be done in various ways, adapted to the problem at hand. The method is thus easy to adapt to many changepoint problems, ranging from univariate to high dimensional. Compared to recently popular random background intervals, seeded intervals lead to reproducibility and much faster computations. For the univariate Gaussian change in mean set-up, the methodology is shown to be asymptotically minimax optimal when paired with appropriate selection criteria. We demonstrate near-linear runtimes and competitive finite sample estimation performance. Furthermore, we illustrate the versatility of our method in high-dimensional settings.

关键词： binary segmentation Breakpoint Fast computation High dimensionality Minimax optimality Multiple changepoint estimation Narrowest-over-threshold method Wild binary segmentation

来源：评论

学校读者我要写书评

暂无评论

Multiple-change-point detection for high dimensional time series via sparsified binary segmentation

引用

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY 2015年第2期77卷 475-507页

作者： Cho, Haeran Fryzlewicz, Piotr Univ Bristol Bristol BS8 1TW Avon England Univ London London Sch Econ & Polit Sci London WC2A 2AE England

Time series segmentation, which is also known as multiple-change-point detection, is a well-established problem. However, few solutions have been designed specifically for high dimensional situations. Our interest is in segmenting the second-order structure of a high dimensional time series. In a generic step of a binary segmentation algorithm for multivariate time series, one natural solution is to combine cumulative sum statistics obtained from local periodograms and cross-periodograms of the components of the input time series. However, the standard maximum' and average' methods for doing so often fail in high dimensions when, for example, the change points are sparse across the panel or the cumulative sum statistics are spuriously large. We propose the sparsified binary segmentation algorithm which aggregates the cumulative sum statistics by adding only those that pass a certain threshold. This sparsifying' step reduces the influence of irrelevant noisy contributions, which is particularly beneficial in high dimensions. To show the consistency of sparsified binary segmentation, we introduce the multivariate locally stationary wavelet model for time series, which is a separate contribution of this work.

关键词： binary segmentation Cumulative sum statistic High dimensional time series Locally stationary wavelet model Multiple-change-point detection Thresholding

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：