Filters








4,343 Hits in 6.6 sec

Research on Parallel Deep Learning for Heterogeneous Computing Architecture

Kaijian Xia, Tao Hu, Wen Si
2020 Journal of Grid Computing  
It is expected that the development and applications of parallel deep learning theories would further influence the field of Heterogeneous Computing.  ...  A typical pattern recognition system is composed of a set of heterogeneous machines, highspeed networks that connect heterogeneous machines, such as a commercialized network or a user specially designed  ... 
doi:10.1007/s10723-020-09520-4 fatcat:m4s4q6v6cjbljjzze4kavuruli

Auto-CNNp: a component-based framework for automating CNN parallelism

Soulaimane Guedria, Noel De Palma, Felix Renard, Nicolas Vuillerme
2019 2019 IEEE International Conference on Big Data (Big Data)  
Auto-CNNp: a componentbased framework for automating CNN parallelism. Abstract-Effectively training of Convolutional Neural Networks (CNNs) is a computationally intensive and time-consuming task.  ...  Therefore, scaling up the training of CNNs has become a key approach to decrease the training duration and train CNN models in a reasonable time.  ...  Experiments presented in this paper were carried out using the Grid'5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as  ... 
doi:10.1109/bigdata47090.2019.9006175 dblp:conf/bigdataconf/GuedriaPRV19 fatcat:7ygej5ryqrh6xb2iqyranly334

GPGPU Accelerated Deep Object Classification on a Heterogeneous Mobile Platform

Syed Rizvi, Gianpiero Cabodi, Denis Patti, Gianluca Francini
2016 Electronics  
Deep convolutional neural networks achieve state-of-the-art performance in image classification.  ...  This paper proposes a complete approach to image classification providing common layers used in neural networks.  ...  Training and testing a neural network is largely a parallel problem, so heterogeneous GPU-CPU architectures are commonly employed to speedup such processes.  ... 
doi:10.3390/electronics5040088 fatcat:oexwebv5wvarlomsmxihvvd2qi

A neural tree model for classification of computing grid resources using PSO tasks scheduling

J. Škrinárová, L. Huraj, V. Siládi
2013 Neural Network World  
This paper proposes a model of neural tree architecture with probabilistic neurons. These trees are used for classification of a large amount of computer grid resources to classes.  ...  Trees are implemented to successfully separate inputs into nine classes of resources. We propose Particle Swarm Optimization model for tasks scheduling in computer grid.  ...  Acknowledgement This work is partially supported by the High Performance Computing Centre of Matej Bel University in Banská Bystrica, Slovakia (HPCC UMB).  ... 
doi:10.14311/nnw.2013.23.014 fatcat:fa72jt33ynap7ewsfzp6lvdmsu

N2Sky - Neural Networks as Services in the Clouds [article]

Erich Schikuta, Erwin Mann
2014 arXiv   pre-print
The system implements a transparent environment aiming to enable both novice and experienced users to do neural network research easily and comfortably.  ...  We present the N2Sky system, which provides a framework for the exchange of neural network specific knowledge, as neural network paradigms and objects, by a virtual organization environment.  ...  This allows the creation of a shared pool of neural net paradigms, neural net objects and other data and information between researchers, developers and end users worldwide. • On-demand self-service.  ... 
arXiv:1401.2468v1 fatcat:lfgxtillkvetfg6xmrfugul6h4

Hardware Accelerated ATLAS Workloads on the WLCG Grid

A C Forti, L Heinrich, M Guth
2020 Journal of Physics, Conference Series  
For this often a large range of network variations must be trained and compared, which for some optimization schemes can be performed in parallel -a workload well suited for Grid computing.  ...  A frequent use-case in the development of machine learning algorithms is the optimization of neural networks through the tuning of their Hyper Parameters (HP).  ...  Figure 2 . 2 Neural net input and job splitting over multiple Grid sites Figure 3 . 3 Parallel coordinates plot for 800 different Hyper Parameter combinations.  ... 
doi:10.1088/1742-6596/1525/1/012086 fatcat:pyojo4sbyjehxe47ior66lmhsa

IAE-Net: Integral Autoencoders for Discretization-Invariant Learning [article]

Yong Zheng Ong and Zuowei Shen and Haizhao Yang
2022 arXiv   pre-print
Compared with alternatives in the literature, IAE-Net achieves state-of-the-art performance in existing applications and creates a wide range of new applications.  ...  This basic building block is applied in parallel in a wide multi-channel structure, which are repeatedly composed to form a deep and densely connected neural network with skip connections as IAE-Net.  ...  This new setting is to simulate the computational environment of many data science problems with heterogeneous training data due to data sensing and collection constraints.  ... 
arXiv:2203.05142v2 fatcat:z3egmt4yybb5ncmgeu7gnzpszu

What can Machine Learning do for Radio Spectrum Management?

Ebtesam Almazrouei, Gabriele Gianini, Nawaf Almoosa, Ernesto Damiani
2020 Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks  
In this work, we discuss the potential bene ts and the challenges with reference to the recent research developments in this area.  ...  The opening of the unlicensed radio spectrum creates new opportunities and new challenges for communication technology that can be faced by Machine Learning techniques.  ...  signals in a harsh environment with low SNR and severe multipath e ect).  ... 
doi:10.1145/3416013.3426443 dblp:conf/mswim/AlmazroueiGAD20 fatcat:2hjionukbjgo5gx6bnknaiymse

Decentralized Asynchronous Learning in Cellular Neural Networks

B. Luitel, G. K. Venayagamoorthy
2012 IEEE Transactions on Neural Networks and Learning Systems  
The current methods involved in learning of CNNs are usually centralized (cells are trained in one location) and synchronous (all cells are trained simultaneously either sequentially or in parallel depending  ...  A decentralized asynchronous learning (DAL) framework for CNN is developed in which each cell of the CNN learns in a spatially and temporally distributed environment.  ...  is implemented in parallel.  ... 
doi:10.1109/tnnls.2012.2216900 pmid:24808070 fatcat:g72ptrutrnekvmdkizs4zosmny

Distributed Deep Learning for Remote Sensing Data Interpretation

J. M. Haut, M. E. Paoletti, S. Moreno, J. Plaza, J. A. Rico, A. Plaza
2022 Zenodo  
in the literature, as well as an exhaustive description of their parallel and distributed implementations (with particular focus on those conducted using cloud computing systems).  ...  We emphasize the role of the cloud in providing a powerful architecture that is now able to manage vast amounts of remotely sensed data due to its implementation simplicity, low cost and high efficiency  ...  current remote sensing applications implemented in cloud environments).  ... 
doi:10.5281/zenodo.6413870 fatcat:c5hea52k2jfebjshcozf5dww7q

IMPLEMENTATION OF DATA MINING TECHNIQUES FOR METEOROLOGICAL APPLICATIONS

A.S. COFIÑO, J.M. GUTIÉRREZ, B. JAKUBIAK, M. MELONEK
2003 Realizing Teracomputing  
One of the main tasks in the Meteorological applications package is the implementation of data mining systems for the analysis of operational and reanalysis databases of atmospheric circulation patterns  ...  This is a more general framework to look at data mining techniques from the meteorological point of view.  ...  Acknowledgements The work described in this paper is supported in part by the European Union through the IST-2001-32243 project "CrossGrid".  ... 
doi:10.1142/9789812704832_0012 fatcat:hzr7wzmetnepxe7yqtbjqmkfti

Distributing SOM Ensemble Training using Grid Middleware

Bogdan L. Vrusias, Leonidas Vomvoridis, Lee Gillam
2007 Neural Networks (IJCNN), International Joint Conference on  
The proposed architecture has been evaluated in a Grid environment, with clock-time performance recorded.  ...  In this paper we explore the distribution of training of self-organised maps (SOM) on Grid middleware.  ...  Grid environment [10] .  ... 
doi:10.1109/ijcnn.2007.4371387 dblp:conf/ijcnn/VrusiasVG07 fatcat:2zbyexdx2bdqlke332y2tlsxqe

RSS-based Indoor Positioning Using Convolutional Neural Network

Safae El Abkari, Abdelilah Jilbab, Jamal El Mhamdi
2020 International Journal of Online and Biomedical Engineering (iJOE)  
We implemented and evaluated our system using a single floor and multi-grid dataset.  ...  Our proposed approach provides a room and grid prediction accuracies of 100% and a mean error of location estimation of 0.98 m.</span></p>  ...  Uses a heterogeneous RBF-neural network and an information fusion algorithm in wireless sensor network. [18] - Uses RS-RBF neural network to fusion monitoring data in an information center.  ... 
doi:10.3991/ijoe.v16i12.16751 fatcat:mqrpcgclnffxrk2ne77rwq6iqi

SHADHO: Massively Scalable Hardware-Aware Distributed Hyperparameter Optimization [article]

Jeff Kinnison, Nathaniel Kremer-Herman, Douglas Thain, Walter Scheirer
2018 arXiv   pre-print
Effectively training these models, however, is not trivial due in part to hyperparameters: user-configured values that control a model's ability to learn from data.  ...  We then conduct model search with SHADHO over the course of one week using 74 GPUs across two compute clusters to optimize U-Net for a cell segmentation task, discovering 515 models that achieve a lower  ...  This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357.  ... 
arXiv:1707.01428v2 fatcat:tyxbvw2vs5hplk722kya7p65za

2021 Index IEEE Transactions on Parallel and Distributed Systems Vol. 32

2022 IEEE Transactions on Parallel and Distributed Systems  
., +, TPDS Aug. 2021 1961-1973 Design and Implementation of a Criticality-and Heterogeneity-Aware Run- time System for Task-Parallel Applications.  ...  Gupta, N., +, TPDS March 2021 575-586 Privacy-Preserving Computation Offloading for Parallel Deep Neural Net- works Training.  ...  Graph coloring Feluca: A Two-Stage Graph Coloring Algorithm With Color-Centric Paradigm on GPU. Zheng, Z., +,  ... 
doi:10.1109/tpds.2021.3107121 fatcat:e7bh2xssazdrjcpgn64mqh4hb4
« Previous Showing results 1 — 15 out of 4,343 results