检索结果-内蒙古大学图书馆

arXiv 2018年

作者： Regier, Jeffrey Pamnany, Kiran Fischer, Keno Noack, Andreas Lam, Maximilian Revels, Jarrett Howard, Steve Giordano, Ryan Schlegel, David McAuliffe, Jon Thomas, Rollin Prabhat Department of Electrical Engineering and Computer Sciences University of California Berkeley United States Parallel Computing Lab Intel Corporation Julia Computing Computer Science and AI Laboratories Massachusetts Institute of Technology Department of Statistics University of California Berkeley United States Lawrence Berkeley National Laboratory

Astronomical catalogs derived from wide-field imaging surveys are an important tool for understanding the Universe. We construct an astronomical catalog from 55 TB of imaging data using Celeste, a Bayesian variational inference code written entirely in the high-productivity programming language Julia. Using over 1.3 million threads on 650,000 Intel Xeon Phi cores of the Cori Phase II supercomputer, Celeste achieves a peak rate of 1.54 DP PFLOP/s. Celeste is able to jointly optimize parameters for 188M stars and galaxies, loading and processing 178 TB across 8192 nodes in 14.6 minutes. To achieve this, Celeste exploits parallelism at multiple levels (cluster, node, and thread) and accelerates I/O through Cori’s Burst Buffer. Julia’s native performance enables Celeste to employ high-level constructs without resorting to handwritten or generated low-level code (C/C++/Fortran), and yet achieve petascale *** Codes 85A35, 68W10, 62P35 Copyright © 2018, The Authors. All rights reserved.

关键词： Inference engines

来源：评论

学校读者我要写书评

暂无评论

Capability Models for Manycore Memory Systems: A Case-Study with Xeon Phi KNL

Capability Models for Manycore Memory Systems: A Case-Study ...

引用

International Symposium on parallel and Distributed Processing (IPDPS)

作者： Sabela Ramos Torsten Hoefler Department of Computer Science Scalable Parallel Computing Lab ETH Zürich

Increasingly complex memory systems and onchip interconnects are developed to mitigate the data movement bottlenecks in manycore processors. One example of such a complex system is the Xeon Phi KNL CPU with three different types of memory, fifteen memory configuration options, and a complex on-chip mesh network connecting up to 72 cores. Users require a detailed understanding of the performance characteristics of the different options to utilize the system efficiently. Unfortunately, peak performance is rarely achievable and achievable performance is hardly documented. We address this with capability models of the memory subsystem, derived by systematic measurements, to guide users to navigate the complex optimization space. As a case study, we provide an extensive model of all memory configuration options for Xeon Phi KNL. We demonstrate how our capability model can be used to automatically derive new close-to-optimal algorithms for various communication functions yielding improvements 5x and 24x over Intel's tuned OpenMP and MPI implementations, respectively. Furthermore, we demonstrate how to use the models to assess how efficiently a bitonic sort application utilizes the memory resources. Interestingly, our capability models predict and explain that the high bandwidthMCDRAM does not improve the bitonic sort performance over DRAM.

关键词： Bandwidth Benchmark testing Random access memory Coherence Program processors Message systems Computational modeling

来源：评论

学校读者我要写书评

暂无评论

Sparse Tensor Factorization on Many-Core Processors with High-Bandwidth Memory

Sparse Tensor Factorization on Many-Core Processors with Hig...

引用

International Symposium on parallel and Distributed Processing (IPDPS)

作者： Shaden Smith Jongsoo Park George Karypis Department of Computer Science and Engineering University of Minnesota Parallel Computing Lab Intel Corporation

HPC systems are increasingly used for data intensive computations which exhibit irregular memory accesses, non-uniform work distributions, large memory footprints, and high memory bandwidth demands. To address these challenging demands, HPC systems are turning to many-core architectures that feature a large number of energy-efficient cores backed by high-bandwidth memory. These features are exemplified in Intel's recent Knights Landing many-core processor (KNL), which typically has 68 cores and 16GB of on-package multi-channel DRAM (MCDRAM). This work investigates how the novel architectural features offered by KNL can be used in the context of decomposing sparse, unstructured tensors using the canonical polyadic decomposition (CPD). The CPD is used extensively to analyze large multi-way datasets arising in various areas including precision healthcare, cybersecurity, and e-commerce. Towards this end, we (i) develop problem decompositions for the CPD which are amenable to hundreds of concurrent threads while maintaining load balance and low synchronization costs; and (ii) explore the utilization of architectural features such as MCDRAM. Using one KNL processor, our algorithm achieves up to 1.8x speedup over a dual socket Intel Xeon system with 44 cores.

关键词： Tensile stress Instruction sets Synchronization Sparse matrices parallel processing Bandwidth

来源：评论

学校读者我要写书评

暂无评论

A semi-supervised method for multi-subject FMRI functional alignment

A semi-supervised method for multi-subject FMRI functional a...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Javier S. Turek Theodore L. Willke Po-Hsuan Chen Peter J. Ramadge Parallel Computing Lab Intel Labs Hillsboro Oregon USA Department of Electrical Engineering Princeton University New Jersey USA

ISBN: (纸本)9781509041183

Practical limitations on the duration of individual fMRI scans have led neuroscientist to consider the aggregation of data from multiple subjects. Differences in anatomical structures and functional topographies of brains require aligning data across subjects. Existing functional alignment methods serve as a preprocessing step that allows subsequent statistical methods to learn from the aggregated multi-subject data. Despite their success, current alignment methods do not leverage the labeled data used in the subsequent methods. In this work we propose a semi-supervised scheme that simultaneously learns the alignment and performs the analysis. We derive a specific instance of the scheme using the Shared Response Model for alignment and Multinomial Logistic Regression for classification. In our experiments this method improves the average classification accuracy from 65.5% to 68.5%, and from 5.3% to 6.1% over the independently-trained methods. Furthermore, our method achieves similar prediction with almost half the samples used for alignment.

关键词： fMRI functional alignment shared response model semi-supervised method

来源：评论

学校读者我要写书评

暂无评论

Translation validation of loop and arithmetic transformations in the presence of recurrences 2016

Translation validation of loop and arithmetic transformation...

引用

17th ACM SIGPLAN/SIGBED Conference on Languages, Compilers, Tools and Theory for Embedded Systems, LCTES 2016

作者： Banerjee, Kunal Mandal, Chittaranjan Sarkar, Dipankar Department of Computer Science and Engineering Indian Institute of Technology Kharagpur India Intel Parallel Computing Lab Bangalore India

ISBN: (纸本)9781450343169

Compiler optimization of array-intensive programs involves extensive application of loop transformations and arithmetic transformations. Hence, translation validation of array-intensive programs requires manipulation of intervals of integers (representing domains of array indices) and relations over such intervals to account for loop transformations and simplification of arithmetic expressions to handle arithmetic transformations. A major obstacle for verification of such programs is posed by the presence of recurrences, whereby an element of an array gets defined in a statement S inside a loop in terms of some other element(s) of the same array which have been previously defined through the same statement S. Recurrences lead to cycles in the data-dependence graph of a program which make dependence analyses and simplifications (through closed-form representations) of the data transformations difficult. Another technique which works better for recurrences does not handle arithmetic transformations. In this work, array data-dependence graphs (ADDGs) are used to represent both the original and the optimized versions of the program and a validation scheme is proposed where the cycles due to recurrences in the ADDGs are suitably abstracted as acyclic subgraphs. Thus, this work provides a unified equivalence checking framework to handle loop and arithmetic transformations along with most of the recurrences - this combination of features had not been achieved by a single verification technique earlier. ©2016 ACM.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Quantum Neuron: An elementary building block for machine learning on quantum computers

arXiv

引用

arXiv 2017年

作者： Cao, Yudong Guerreschi, Gian Giacomo Aspuru-Guzik, Alán Department of Chemistry and Chemical Biology Harvard University CambridgeMA02138 United States Parallel Computing Lab Intel Corporation Santa ClaraCA95054 Canadian Institute for Advanced Research TorontoONM5G 1Z8 Canada

Even the most sophisticated artificial neural networks are built by aggregating substantially identical units called neurons. A neuron receives multiple signals, internally combines them, and applies a non-linear function to the resulting weighted sum. Several attempts to generalize neurons to the quantum regime have been proposed, but all proposals collided with the difficulty of implementing non-linear activation functions, which is essential for classical neurons, due to the linear nature of quantum mechanics. Here we propose a solution to this roadblock in the form of a small quantum circuit that naturally simulates neurons with threshold activation. Our quantum circuit defines a building block, the "quantum neuron", that can reproduce a variety of classical neural network constructions while maintaining the ability to process superpositions of inputs and preserve quantum coherence and entanglement. In the construction of feedforward networks of quantum neurons, we provide numerical evidence that the network not only can learn a function when trained with superposition of inputs and the corresponding output, but that this training suffices to learn the function on all individual inputs separately. When arranged to mimic Hopfield networks, quantum neural networks exhibit properties of associative memory. Patterns are encoded using the simple Hebbian rule for the weights and we demonstrate attractor dynamics from corrupted inputs. Finally, the fact that our quantum model closely captures (traditional) neural network dynamics implies that the vast body of literature and results on neural networks becomes directly relevant in the context of quantum machine learning. Copyright © 2017, The Authors. All rights reserved.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

RedLib: Nestable Reductions for Collections in Java

RedLib: Nestable Reductions for Collections in Java

引用

IEEE International Conference on High Performance computing and Communications

作者： Mostafa Mehrabi Xing Fan Nasser Giacaman Oliver Sinnen Department of Electrical and Computer Engineering The University of Auckland Parallel and Reconfigurable Computing Lab

ISBN: (纸本)9781509042982

A reduction is a parallel programming mechanism for combining two or more elements into one. Many parallel programming languages, tools and frameworks (e.g., OpenMP, MPI, etc.) directly support simple forms of reductions (e.g., building a total sum out of partial sums). Some of those tools and frameworks allow more complex reductions to be implemented as custom reductions. However, the success of network-based application frameworks like Hadoop have shown that there is a strong need for reductions of aggregate data structures, such as the union of sets or maps. Usually parallel programming frameworks on shared-memory systems do not support these types of complex reductions directly, and a user needs to implement them manually. To address the gap thereof, this paper proposes an object-oriented reduction framework that supports reductions of aggregate types, and proposes the nesting of reduction objects for flexible extensions of reduction operations. Based on the proposed framework, a reduction library (RedLib) has been developed for Java with direct support for many common reduction operations on collections and maps. Furthermore, the paper studies the usage of the framework for common and complex cases and evaluates its performance, based on operations found in standard benchmarks.

关键词： object oriented reductions shared memory reductions nestable reductions aggregate reductions collections maps :

来源：评论

学校读者我要写书评

暂无评论

Controlled vs. Automatic Processing: A Graph-Theoretic Approach to the Analysis of Serial vs. parallel Processing in Neural Network Architectures 38

Controlled vs. Automatic Processing: A Graph-Theoretic Appro...

引用

38th Annual Meeting of the Cognitive Science Society: Recognizing and Representing Events, CogSci 2016

作者： Musslick, Sebastian Dey, Biswadip Özcimder, Kayhan Patwary, Md. Mostofa Ali Willke, Theodore L. Cohen, Jonathan D. Princeton Neuroscience Institute Princeton University PrincetonNJ08544 United States Department of Mechanical and Aerospace Engineering Princeton University PrincetonNJ08544 United States Parallel Computing Lab Intel Corporation Santa ClaraCA95054 United States

ISBN: (纸本)9780991196739

The limited ability to simultaneously perform multiple tasks is one of the most salient features of human performance and a defining characteristic of controlled processing. Based on the assumption that multitasking constraints arise from shared representations between individual tasks, we describe a graph-theoretic approach to analyze these constraints. Our results are consistent with previous numerical work (Feng, Schwemmer, Gershman, & Cohen, 2014), showing that even modest amounts of shared representation induce dramatic constraints on the parallel processing capability of a network architecture. We further illustrate how this analysis method can be applied to specific neural networks to efficiently characterize the full profile of their parallel processing capabilities. We present simulation results that validate theoretical predictions, and discuss how these methods can be applied to empirical studies of controlled vs. and automatic processing and multitasking performance in humans. © 2016 Proceedings of the 38th Annual Meeting of the Cognitive Science Society, CogSci 2016. All rights reserved.

关键词： Multitasking

来源：评论

学校读者我要写书评

暂无评论

Real-time full correlation matrix analysis of fMRI data

Real-time full correlation matrix analysis of fMRI data

引用

IEEE International Conference on Big Data

作者： Yida Wang Bryn Keller Mihai Capota Michael J. Anderson Narayanan Sundaram Jonathan D. Cohen Kai Li Nicholas B. Turk-Browne Theodore L. Willke Parallel Computing Lab Intel Corporation Princeton Neuroscience Institute Princeton University Department of Computer Science Princeton University

ISBN: (纸本)9781467390064

Real-time functional magnetic resonance imaging (rtfMRI) is an emerging approach for studying the functioning of the human brain. Computational challenges combined with high data velocity have to this point restricted rtfMRI analyses to studying regions of the brain independently. However, given that neural processing is accomplished via functional interactions among brain regions, neuroscience could stand to benefit from rtfMRI analyses of full-brain interactions. In this paper, we extend such an offline analysis method, full correlation matrix analysis (FCMA), to enable its use in rtfMRI studies. Specifically, we introduce algorithms capable of processing real-time data for all stages of the FCMA machine learning workflow: incremental feature selection, model updating, and real-time classification. We also present an actor-model based distributed system designed to support FCMA and other rtfMRI analysis methods. Experiments show that our system successfully analyzes a stream of brain volumes and returns neurofeedback with less than 180 ms of lag. Our real-time FCMA implementation provides the same accuracy as an optimized offline FCMA toolbox while running 3.6-6.2× faster.

关键词： Big data Machine-to-machine communications Conferences

来源：评论

学校读者我要写书评

暂无评论

Using behavior to decode allocation of attention in context dependent decision making 14

Using behavior to decode allocation of attention in context ...

引用

14th International Conference on Cognitive Modeling, ICCM 2016

作者： Shvartsman, Michael Srivastava, Vaibhav Sundaram, Narayanan Cohen, Jonathan D. Princeton Neuroscience Institute Washington Rd. PrincetonNJ08544 United States Department of Mechanical and Aerospace Engineering 41 Olden St. PrincetonNJ08544 United States Intel Parallel Computing Lab 2200 Mission College Blvd. Santa ClaraCA95054 United States

ISBN: (纸本)9780998508207

We present a model of the dynamics of adaptive attention allocation in the AX Continuous Performance Test (AX-CPT), a simple context dependent decision making task of interest to the research communities concerned with cognitive control, schizophrenia, anxiety and aging (Braver et al., 2001;Cohen et al., 1999;Eysenck et al., 2007). We ground it in our recent theory of decision making under dynamic context, that assumes humans use sequential Bayesian inference to combine information from multiple sources in perception and (optionally) memory over time. The theory generalizes the well-known diffusion decision model of single-stimulus decision making (DDM;Ratcliff, 1978). Our first result is a new analysis that shows how memory encoding and retention can yield a variable initial condition for either a single- or multiple-stimulus decision, providing a theoretical grounding to the assumption of variability in initial condition previously shown to improve data fits for the DDM. Our second result is using this model to decode attention allocation from behavioral data in a novel quantitative payoff manipulation in the AX-CPT, showing that our model can capture the differences in how subjects encode and retrieve contextual information when the relative emphasis on task speed and accuracy is changed. Copyright © TRECVID *** rights reserved.

关键词： Decision making

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：