检索结果-内蒙古大学图书馆

Fault Tolerance for HPC at eXtreme Scale (FTXS)

作者： Yehonatan Fridman Yaniv Snir Harel Levin Danny Hendler Hagit Attiya Gal Oren Department of Computer Science Ben-Gurion University of the Negev Israel Atomic Energy Commission Google Mobileye Vision Technologies Scientific Computing Center Nuclear Research Center – Negev Department of Computer Science Technion – Israel Institute of Technology

ISBN: (纸本)9781665488488

HPC systems are a critical resource for scientific research and advanced industries. The increased demand for computational power and memory ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of numerous compute nodes and are consequently expected to experience frequent faults and *** solvers, in particular, iterative linear solvers are key building block in numerous large-scale scientific applications. Consequently, supporting the recovery of distributed solvers is necessary for scaling scientific applications to exascale platforms. Previous recovery methods for iterative solvers are based on Checkpoint-Restart (CR), which incurs high fault tolerance overhead, or intrinsic fault tolerance, which require extra computation time to converge after *** state reconstruction (ESR) was proposed as an alternative mechanism to alleviate the impact of frequent failures on long-term computations. ESR has been shown to provide exact reconstruction of the computation state while avoiding the need for costly checkpointing. However, ESR currently relies on volatile memory for fault tolerance, and must therefore maintain redundancies in the RAM of multiple nodes. This not only incurs high memory overhead but also prevents ESR from being fully resilient, that is, resilient against a full system *** supercomputer designs feature emerging non-volatile RAM (NVRAM) technology, for example, the exascale Aurora that is planned to consist of Intel Optane™ DCPMM. This paper investigates how NVRAM can be utilized to devise an enhanced ESR-based recovery mechanism that is more efficient and provides full resilience. Our mechanism, called in-NVRAM ESR, provides full resiliency while significantly reducing both the memory footprint and the time overhead in comparison with the original ESR design (in-RAM ESR). In-NVRAM ESR is based on a novel MPI One-Si

关键词： Nonvolatile memory Source coding Fault tolerant systems Redundancy Random access memory Supercomputers computer crashes

来源：评论

学校读者我要写书评

暂无评论

Fast Dynamic Walking with RH5 Humanoid Robot

Fast Dynamic Walking with RH5 Humanoid Robot

引用

IEEE-RAS International Conference on Humanoid Robots

作者： Ivan Bergonzani Mihaela Popescu Shivesh Kumar Frank Kirchner Robotics Innovation Center German Research Center for Artificial Intelligence (DFKI GmbH) Bremen Germany Robotics Group Faculty of Mathematics and Computer Science University of Bremen Bremen Germany Department of Mechanics & Maritime Sciences Division of Dynamics Chalmers University of Technology Gothenburg Sweden

Humanoid robots have the potential of becoming general purpose robots augmenting the human work-force in industries. However, they must match the agility and versatility of humans. It is particularly challenging for humanoids actuated with electric drives to achieve that as one must strive for the right balance between mass-inertial distribution in the robot as well as velocity and force transmissions in its actuation concept. In addition to optimal design of the robot, the control system must be designed to exploit the full potential of the robot. In this paper, we perform experimental investigations on the dynamic walking capabilities of a series-parallel hybrid humanoid named RH5. We demonstrate that it is possible to walk up to speeds of 0.43 m/s with a position controlled robot without full state feedback which makes it one of the fastest walking humanoids with similar size and actuation modalities. Video of the experiments is available at: https://***/39GL2vPedGY

关键词：

来源：评论

学校读者我要写书评

暂无评论

GEM: Context-Aware Gaze EstiMation with Visual Search Behavior Matching for Chest Radiograph

arXiv

引用

arXiv 2024年

作者： Liu, Shaonan Chen, Wenting Liu, Jie Luo, Xiaoling Shen, Linlin Computer Vision Institute College of Computer Science and Software Engineering Shenzhen University China Department of Electrical Engineering City University of Hong Kong Hong Kong AI Research Center for Medical Image Analysis and Diagnosis Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

Gaze estimation is pivotal in human scene comprehension tasks, particularly in medical diagnostic analysis. Eye-tracking technology facilitates the recording of physicians’ ocular movements during image interpretation, thereby elucidating their visual attention patterns and information-processing strategies. In this paper, we initially define the context-aware gaze estimation problem in medical radiology report settings. To understand the attention allocation and cognitive behavior of radiologists during the medical image interpretation process, we propose a context-aware Gaze EstiMation (GEM) network that utilizes eye gaze data collected from radiologists to simulate their visual search behavior patterns throughout the image interpretation process. It consists of a context-awareness module, visual behavior graph construction, and visual behavior matching. Within the context-awareness module, we achieve intricate multimodal registration by establishing connections between medical reports and images. Subsequently, for a more accurate simulation of genuine visual search behavior patterns, we introduce a visual behavior graph structure, capturing such behavior through high-order relationships (edges) between gaze points (nodes). To maintain the authenticity of visual behavior, we devise a visual behavior-matching approach, adjusting the high-order relationships between them by matching the graph constructed from real and estimated gaze points. Extensive experiments on four publicly available datasets demonstrate the superiority of GEM over existing methods and its strong generalizability, which also provides a new direction for the effective utilization of diverse modalities in medical image interpretation and enhances the interpretability of models in the field of medical imaging. https://***/Tiger-SN/GEM. Copyright © 2024, The Authors. All rights reserved.

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Exploring Parameter-Efficient Fine-Tuning to Enable Foundation Models in Federated Learning

arXiv

引用

arXiv 2022年

作者： Sun, Guangyu Khalid, Umar Mendieta, Matias Wang, Pu Chen, Chen Center for Research in Computer Vision University of Central Florida OrlandoFL United States Department of Computer Science University of North Carolina at Charlotte CharlotteNC United States

Federated learning (FL) has emerged as a promising paradigm for enabling the collaborative training of models without centralized access to the raw data on local devices. In the typical FL paradigm (e.g., FedAvg), model weights are sent to and from the server each round to participating clients. Recently, the use of small pre-trained models has been shown to be effective in federated learning optimization and improving convergence. However, recent state-of-the-art pre-trained models are getting more capable but also have more parameters, known as the "Foundation Models." In conventional FL, sharing the enormous model weights can quickly put a massive communication burden on the system, especially if more capable models are employed. Can we find a solution to enable those strong and readily available pre-trained models in FL to achieve excellent performance while simultaneously reducing the communication burden? To this end, we investigate the use of parameter-efficient fine-tuning in federated learning and thus introduce a new framework: FedPEFT. Specifically, we systemically evaluate the performance of FedPEFT across a variety of client stability, data distribution, and differential privacy settings. By only locally tuning and globally sharing a small portion of the model weights, significant reductions in the total communication overhead can be achieved while maintaining competitive or even better performance in a wide range of federated learning scenarios, providing insight into a new paradigm for practical and effective federated systems. Copyright © 2022, The Authors. All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Recovery of Distributed Iterative Solvers for Linear Systems Using Non-Volatile RAM

arXiv

引用

arXiv 2022年

作者： Fridman, Yehonatan Snir, Yaniv Levin, Harel Hendler, Danny Attiya, Hagit Oren, Gal Department of Computer Science Ben-Gurion University of the Negev Israel Atomic Energy Commission Google Mobileye Vision Technologies Scientific Computing Center Nuclear Research Center - Negev Department of Computer Science Technion - Israel Institute of Technology

HPC systems are a critical resource for scientific research and advanced industries. The increased demand for computational power and memory ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of numerous compute nodes and are consequently expected to experience frequent faults and crashes. Mathematical solvers, in particular, iterative linear solvers are key building block in numerous large-scale scientific applications. Consequently, supporting the recovery of distributed solvers is necessary for scaling scientific applications to exascale platforms. Previous recovery methods for iterative solvers are based on Checkpoint-Restart (CR), which incurs high fault tolerance overhead, or intrinsic fault tolerance, which require extra computation time to converge after failures. Exact state reconstruction (ESR) was proposed as an alternative mechanism to alleviate the impact of frequent failures on long-term computations. ESR has been shown to provide exact reconstruction of the computation state while avoiding the need for costly checkpointing. However, ESR currently relies on volatile memory for fault tolerance, and must therefore maintain redundancies in the RAM of multiple nodes. This not only incurs high memory overhead but also prevents ESR from being fully resilient, that is, resilient against a full system crash. Recent supercomputer designs feature emerging non-volatile RAM (NVRAM) technology, for example, the exascale Aurora that is planned to consist of Intel Optane™ DCPMM. This paper investigates how NVRAM can be utilized to devise an enhanced ESR-based recovery mechanism that is more efficient and provides full resilience. Our mechanism, called in-NVRAM ESR, provides full resiliency while significantly reducing both the memory footprint and the time overhead in comparison with the original ESR design (in-RAM ESR). In-NVRAM ESR is based on a novel MPI One

关键词： Random access storage

来源：评论

学校读者我要写书评

暂无评论

Enhancing UAV Autonomous Navigation in Indoor Environments Using Reinforcement Learning and Convolutional Neural Networks 22

Enhancing UAV Autonomous Navigation in Indoor Environments U...

引用

22nd IEEE International Symposium on Intelligent Systems and Informatics, SISY 2024

作者： Safarpour Motealegh Mahalegi, Homayoun Farhadi, Amirfarhad Molnar, Gyorgy Nagy, Eniko Obuda University Institute of Mechatronics and Vehicle Engineering Budapest Hungary Islamic Azad University Department of Computer Engineering Science and Research Branch Tehran Iran Obuda University Hungary Szechenyi University Apaczai Csere János Faculty of Humanities Education and Social Sciences Kando Kalman Faculty of Electrical Engineering Budapest Gyor Hungary Obuda University Antal Bejczy Center for Intelligent Robotics John von Neumann Faculty of Informatics Budapest Hungary

ISBN: (纸本)9798350385601

In recent years, the number of unmanned aerial vehicle (UAV) applications has increased. However, navigating them indoors is still tricky because no GPS signals are available, and the obstacles constantly change. This study exploits the collaboration between convolutional neural networks (CNNs) and reinforcement learning (RL) to overcome these challenges, improving the level of independence of UAVs in indoor settings. We present a new technique known as potential-based reward shaping, which directly incorporates expert knowledge into the reinforcement learning framework. This technique efficiently addresses the problem of sparse rewards that often hinder learning in conventional RL configurations. In addition, we have created a reflective reinforcement learning agent that systematically assesses its actions to prioritize those that result in the most substantial enhancements. We have extensively evaluated the performance of our method with the High-Definition Indoor Navigation (HDIN) dataset to show that our method continuously outperformed current non-introspective RL techniques in terms of decision-making speed and navigation accuracy in this evaluation. The results show that this new CNN-RL combination is feasible and suggests a scalable method for complex indoor navigation problems. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

FogROS2-FT: Fault Tolerant Cloud robotics

FogROS2-FT: Fault Tolerant Cloud Robotics

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Kaiyuan Chen Kush Hari Trinity Chung Michael Wang Nan Tian Christian Juette Jeffrey Ichnowski Liu Ren John Kubiatowicz Ion Stoica Ken Goldberg Department of Electrical Engineering and Computer Science University of California Berkeley CA USA Robert Bosch Research and Technology Center North America Sunnyvale CA USA Robotics Institute Carnegie Mellon University Department of Industrial Engineering and Operations Research University of California Berkeley CA USA

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Cloud robotics enables robots to offload complex computational tasks to cloud servers for performance and ease of management. However, cloud compute can be costly, cloud services can suffer occasional downtime, and connectivity between the robot and cloud can be prone to variations in network Quality-of-Service (QoS). We present FogROS2-FT (Fault Tolerant) to mitigate these issues by introducing a multi-cloud extension that automatically replicates independent stateless robotic services, routes requests to these replicas, and directs the first response back. With replication, robots can still benefit from cloud computations even when a cloud service provider is down or there is low QoS. Additionally, many cloud computing providers offer low-cost "spot" computing instances that may shutdown unpredictably. Normally, these low-cost instances would be inappropriate for cloud robotics, but the fault tolerance nature of FogROS2-FT allows them to be used reliably. We demonstrate FogROS2-FT fault tolerance capabilities in 3 cloud-robotics scenarios in simulation (visual object detection, semantic segmentation, motion planning) and 1 physical robot experiment (scan-pick-and-place). Running on the same hardware specification, FogROS2-FT achieves motion planning with up to 2.2x cost reduction and up to a 5.53x reduction on 99 Percentile (P99) long-tail latency. FogROS2-FT reduces the P99 long-tail latency of object detection and semantic segmentation by 2.0x and 2.1x, respectively, under network slowdown and resource contention. Videos and code are available at https://***/view/fogros2-ft.

关键词： Fault tolerance Heavily-tailed distribution Semantic segmentation Fault tolerant systems Quality of service Object detection Planning Servers Robots Cloud robotics

来源：评论

学校读者我要写书评

暂无评论

St-Keys: Self-Supervised Transformer for Keyword Spotting in Historical Handwritten Documents

SSRN

引用

SSRN 2023年

作者： Jemni, Sana Khamekhem Ammar, Sourour Souibgui, Mohamed Ali Kessentini, Yousri Cheddad, Abbas Digital Research Center of Sfax B.P. 275 Sakiet Ezzit Sfax3021 Tunisia SM@RTS: Laboratory of Signals systeMs aRtificial Intelligence and neTworkS MIR@CL: Multimedia InfoRmation systems and Advanced Computing Laboratory Computer Vision Center Computer Science Department Universitat Aut`onoma de Barcelona Spain Department of Computer Science Blekinge Institute of Technology Karlskrona Sweden

Keyword spotting (KWS) in historical documents is an important tool for the initial exploration of digitized collections. Nowadays, the most efficient KWS methods are relying on machine learning techniques that require a large amount of annotated training data. However, in the case of historical manuscripts, there is a lack of annotated corpus for training. To handle the data scarcity issue, we investigate the merits of the self-supervised learning to extract useful representations of the input data without relying on human annotations and then using these representations in the downstream task. We propose ST-KeyS, a masked auto-encoder model based on vision transformers where the pretraining stage is based on the mask-and-predict paradigm, without the need of labeled data. In the fine-tuning stage, the pre-trained encoder is integrated into a siamese neural network model that is fine-tuned to improve feature embedding from the input images. We further improve the image representation using pyramidal histogram of characters (PHOC) embedding to create and exploit an intermediate representation of images based on text attributes. In an exhaustive experimental evaluationon three widely used benchmark datasets (Botany, Alvermann Konzilsprotokolle and George Washington), the proposed approach outperforms state-of-the-art methods trained on the same datasets. © 2023, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

ST-KeyS: Self-Supervised Transformer for Keyword Spotting in Historical Handwritten Documents

arXiv

引用

arXiv 2023年

作者： Jemni, Sana Khamekhem Ammar, Sourour Souibgui, Mohamed Ali Kessentini, Yousri Cheddad, Abbas Digital Research Center of Sfax B.P. 275 Sakiet Ezzit Sfax3021 Tunisia Laboratory of Signals systems Artificial Intelligence and networks Sfax Tunisia Multimedia Information systems and Advanced Computing Laboratory Tunisia Computer Vision Center Computer Science Department Universitat Autònoma de Barcelona Spain Department of Computer Science Blekinge Institute of Technology Karlskrona Sweden

Keyword spotting (KWS) in historical documents is an important tool for the initial exploration of digitized collections. Nowadays, the most efficient KWS methods are relying on machine learning techniques that require a large amount of annotated training data. However, in the case of historical manuscripts, there is a lack of annotated corpus for training. To handle the data scarcity issue, we investigate the merits of the self-supervised learning to extract useful representations of the input data without relying on human annotations and then using these representations in the downstream task. We propose ST-KeyS, a masked auto-encoder model based on vision transformers where the pretraining stage is based on the mask-and-predict paradigm, without the need of labeled data. In the fine-tuning stage, the pre-trained encoder is integrated into a siamese neural network model that is fine-tuned to improve feature embedding from the input images. We further improve the image representation using pyramidal histogram of characters (PHOC) embedding to create and exploit an intermediate representation of images based on text attributes. In an exhaustive experimental evaluation on three widely used benchmark datasets (Botany, Alvermann Konzilsprotokolle and George Washington), the proposed approach outperforms state-of-the-art methods trained on the same datasets. © 2023, CC BY.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Compressing deep cnns using basis representation and spectral fine-tuning

arXiv

引用

arXiv 2021年

作者： Tayab, Muhammad Khan, Fahad Ahmad Mahalanobis, Abhijit Center for Research in Computer Vision Department of Computer Science University of Central Florida United States

We propose an efficient and straightforward method for compressing deep convolutional neural networks (CNNs) that uses basis filters to represent the convolutional layers, and optimizes the performance of the compressed network directly in the basis space. Specifically, any spatial convolution layer of the CNN can be replaced by two successive convolution layers: the first is a set of three-dimensional orthonormal basis filters, followed by a layer of one-dimensional filters that represents the original spatial filters in the basis space. We jointly fine-tune both the basis and the filter representation to directly mitigate any performance loss due to the truncation. Generality of the proposed approach is demonstrated by applying it to several well known deep CNN architectures and data sets for image classification and object detection. We also present the execution time and power usage at different compression levels on the Xavier Jetson AGX processor. © 2021, CC BY.

关键词： Convolution

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：