4,208 Hits in 6.4 sec

Distributed Fusion of Heterogeneous Remote Sensing and Social Media Data: A Review and New Developments

Jun Li, Zhenjie Liu, Xinya Lei, Lizhe Wang
2021 Proceedings of the IEEE  
A new distributed fusion framework that can accelerate the fusion of heterogeneous remote sensing and social media data is proposed by decomposing large data sets into small ones and processing them in  ...  real-time response is needed.  ...  C A S E S T U D Y: H E T E R O G E N E O U S D A T A F U S I O N U S I N G D I S T R I B U T E D C O M P U T I N G As a follow-up to Section III, we discuss a case study related to a flood event in Boulder  ... 
doi:10.1109/jproc.2021.3079176 fatcat:gk2xqgsipjfr7kfanauymtk724

Preparing Nuclear Astrophysics for Exascale [article]

Max P. Katz, Ann Almgren, Maria Barrios Sazo, Kiran Eiden, Kevin Gott, Alice Harpole, Jean M. Sexton, Don E. Willcox, Weiqun Zhang, Michael Zingale
2020 arXiv   pre-print
Castro and MAESTROeX are nuclear astrophysics codes that simulate thermonuclear fusion in the context of supernovae and X-ray bursts.  ...  In this paper we describe the changes that have been made to these codes to transform them from standard MPI + OpenMP codes targeted at petascale CPU-based systems into a form compatible with the pre-exascale  ...  The loss in efficiency at higher scales occurs because the code is spending more time in message passing (as MPI ranks send their data to other MPI ranks that neighbor them in the simulation domain).  ... 
arXiv:2007.05218v1 fatcat:x32sereeenf5niid65px6i5cci

A Unified Optimization Approach for Sparse Tensor Operations on GPUs

Bangtian Liu, Chengyao Wen, Anand D. Sarwate, Maryam Mehri Dehnavi
2017 2017 IEEE International Conference on Cluster Computing (CLUSTER)  
The performance of the proposed unified approach is demonstrated for tensor-based kernels such as the Sparse Matricized Tensor- Times-Khatri-Rao Product (SpMTTKRP) and the Sparse Tensor- Times-Matrix Multiply  ...  We implement a CANDECOMP/PARAFAC (CP) decomposition and achieve up to 14.9 times speedup using the unified method over state-of-the-art libraries on NVIDIA Titan-X GPUs.  ...  As shown in the Figure 8 , when the rank varies from 8 to 64, the execution time of ParTI increases at a faster rate compared to unified.  ... 
doi:10.1109/cluster.2017.75 dblp:conf/cluster/LiuWSD17 fatcat:meo46pqswreuvkw3rcdm444bgy

Breaking the Computation and Communication Abstraction Barrier in Distributed Machine Learning Workloads [article]

Abhinav Jangda, Jun Huang, Guodong Liu, Amir Hossein Nodehi Sabet, Saeed Maleki, Youshan Miao, Madanlal Musuvathi, Todd Mytkowicz, Olli Sarikivi
2022 arXiv   pre-print
Manually applying these optimizations needs modifications in underlying computation and communication libraries for each scenario, which is time consuming and error-prone.  ...  However, current logical separation between computation and communication kernels in deep learning frameworks misses the optimization opportunities across such barrier.  ...  Unifying the expression of computation and communication for distributed deep learning in the same DSL is the foundation to enable optimizations across computation and communication.  ... 
arXiv:2105.05720v5 fatcat:qg5o27bgljbi3eyvthreygrzgu

SIMD-X: Programming and Processing of Graph Algorithms on GPUs [article]

Hang Liu, H. Howie Huang
2018 arXiv   pre-print
In addition, SIMD-X leverages push-pull based kernel fusion that, with the help of a new deadlock-free global barrier, reduces a large number of computation kernels to very few.  ...  To this end, SIMD-X utilizes just-in-time task management which filters out inactive vertices at runtime and intelligently maps various tasks to different amount of GPU cores in pursuit of workload balancing  ...  For instance, "think like a graph" [58] requires each vertex to obtain the view of the entire partition on one machine in order to minimize the communication cost.  ... 
arXiv:1812.04070v1 fatcat:b6gbadwsmnfpvao3fley2c7xty

An Induced Multi-Relational Framework for Answer Selection in Community Question Answer Platforms [article]

Kanika Narang, Chaoqi Yang, Adit Krishnan, Junting Wang, Hari Sundaram, Carolyn Sutter
2019 arXiv   pre-print
Third, we show a surprising result---boosting techniques improve learning over familiar stacking, fusion, or aggregation approaches for neural architectures.  ...  We develop a novel induced relational graph convolutional network (IR-GCN) framework to address the question. We make three contributions.  ...  The key question that follows is, how do we combine these diverse views in a unified learning framework?  ... 
arXiv:1911.06957v1 fatcat:ijh5d4z76vgg5j7qqkhw2d6i24

Towards hierarchical context: unfolding visual community potential for interactive video retrieval

Lin Pang, Juan Cao, Lei Bao, Yongdong Zhang, Shouxun Lin
2010 Multimedia tools and applications  
In this paper, we exploit the visual community structure in visual-temporal correlation network and utilize it to improve interactive video retrieval.  ...  Firstly, we propose a hierarchical community-based feedback algorithm.  ...  Thus, in this work, we aim to introduce a unified framework combining both feedback algorithm and visualization interface.  ... 
doi:10.1007/s11042-010-0605-0 fatcat:vod2kc5gavhe7ei2e4dmzdubg4

Digital Twin Technology [chapter]

Zongyan Wang
2020 Industry 4.0 - Impact on Intelligent Logistics and Manufacturing  
It is the mapping technology for the whole lifecycle process of physical equipment in virtual space. It is the basic technology of Industrial 4.0.  ...  This chapter mainly introduces: (1) the generation of digital twin technology; (2) the definition and characteristics of digital twin technology; (3) the relationship between digital twin and digital thread  ...  They did a lot of work for this chapter, and a special thanks for their efforts.  ... 
doi:10.5772/intechopen.80974 fatcat:tqw26k47kffj5byikjcxsdgvnm

BlueFog: Make Decentralized Algorithms Practical for Optimization and Deep Learning [article]

Bicheng Ying, Kun Yuan, Hanbin Hu, Yiming Chen, Wotao Yin
2021 arXiv   pre-print
Based on a unified abstraction of various communication operations, BlueFog offers intuitive interfaces to implement a spectrum of decentralized algorithms, from those using a static, undirected graph  ...  Decentralized algorithm is a form of computation that achieves a global goal through local dynamics that relies on low-cost communication between directly-connected agents.  ...  II-A for the discussion on the weight matrix. 3) Communicate in a synchronous or asynchronous mode.  ... 
arXiv:2111.04287v1 fatcat:ei7xa3r6czfnvhyglt42sd3dca

GC3: An Optimizing Compiler for GPU Collective Communication [article]

Meghan Cowan, Saeed Maleki, Madanlal Musuvathi, Olli Saarikivi, Yifan Xiong
2022 arXiv   pre-print
As models grow in size and execute on more GPUs, the collective communications used in these applications become a bottleneck.  ...  This paper introduces GC3, a system for programmable GPU communication.  ...  This paper proposes GC3, a unified framework that provides both algorithmic flexibility and performance.  ... 
arXiv:2201.11840v3 fatcat:542qd5tmozb3vphtjplgshjmiu

Introduction to the Special Issue on Tensor Decomposition for Signal Processing and Machine Learning

Hongyang Chen, Sergiy A. Vorobyov, Hing Cheung So, Fauzia Ahmad, Fatih Porikli
2021 IEEE Journal on Selected Topics in Signal Processing  
The identified drug compounds were mainly related to known antiviral drugs, several of which were also previously identified via in silico experiments for treating COVID-19.  ...  introduced, and then is combined with a new pruning method for reducing the search space, and multi-thread optimization to further improve the execution performance.  ... 
doi:10.1109/jstsp.2021.3065184 fatcat:qbvihejwkfaa5hoztety77pnwi

Performance Portability of HPC Discovery Science Software: Fusion Energy Turbulence Simulations at Extreme Scale

2017 Supercomputing Frontiers and Innovations  
Important application domains, such as Magnetic Fusion Energy (MFE), have improved modeling of increasingly complex physical systemsespecially with respect to reducing "time-to-solution" as well as "energy  ...  of performance scaling and code portability for path-to-exascale platforms.  ...  For other systems, assigning consecutive ranks for processes within each toroidal communicator generally leads to improved performance.  ... 
doi:10.14529/jsfi170105 fatcat:geqrfzmrrze27plhvp54fzqbka

Elastography Using Multi-Stream GPU: An Application to Online Tracked Ultrasound Elastography, In-Vivo and the da Vinci Surgical System

Nishikant P. Deshmukh, Hyun Jae Kang, Seth D. Billings, Russell H. Taylor, Gregory D. Hager, Emad M. Boctor, Assad Anshuman Oberai
2014 PLoS ONE  
A system for real-time ultrasound (US) elastography will advance interventions for the diagnosis and treatment of cancer by advancing methods such as thermal monitoring of tissue ablation.  ...  Since EM tracking cannot be used in all systems, an integration of real-time elastography and the da Vinci Surgical System is presented and evaluated for elastography stream quality based on our metric  ...  Pezhman Foroughi for his valuable input and source code of TRuE, Dr. Daniel Carnegie for help with the animal experiments, and Dr. Ioana Flemings, Alexis Cheng, and Xiayuo Gao for valuable inputs.  ... 
doi:10.1371/journal.pone.0115881 pmid:25541954 pmcid:PMC4277422 fatcat:5bwsbqcp4ja7bh3fpp3ddhp7wq

BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation [article]

Zhijian Liu, Haotian Tang, Alexander Amini, Xinyu Yang, Huizi Mao, Daniela Rus, Song Han
2022 arXiv   pre-print
In this paper, we break this deeply-rooted convention with BEVFusion, an efficient and generic multi-task multi-sensor fusion framework.  ...  Multi-sensor fusion is essential for an accurate and reliable autonomous driving system. Recent approaches are based on point-level fusion: augmenting the LiDAR point cloud with camera features.  ...  We would like to thank Xuanyao Chen and Brady Zhou for their guidance on detection and segmentation evaluation, and Yingfei Liu and Tiancai Wang for their helpful discussions.  ... 
arXiv:2205.13542v2 fatcat:qtunylgozjcvrdrjzdk23xjpve

A Proposed Conceptual Framework for a Representational Approach to Information Retrieval [article]

Jimmy Lin
2021 arXiv   pre-print
I show that many recently proposed retrieval methods, including multi-stage ranking designs, can be seen as different parameterizations in this framework, and that a unified view suggests a number of open  ...  This paper outlines a conceptual framework for understanding recent developments in information retrieval and natural language processing that attempts to integrate dense and sparse retrieval methods.  ...  Acknowledgements This research was supported in part by the Natural Sciences and Engineering Research Council (NSERC) of Canada.  ... 
arXiv:2110.01529v2 fatcat:iluzpawvjbbwdan3ei2aanywm4
« Previous Showing results 1 — 15 out of 4,208 results