Filters








14,500 Hits in 5.0 sec

Blind Optimization for Exploiting Hardware Features [chapter]

Dan Knights, Todd Mytkowicz, Peter F. Sweeney, Michael C. Mozer, Amer Diwan
2009 Lecture Notes in Computer Science  
Second, hardware manufacturers do not reveal all details of their microprocessors so even if the authors of optimizations wanted to simultaneously optimize for all components of the hardware, they may  ...  Blind optimization uses the insight that we can generate many variants of an application by altering semantic preserving parameters of an application; for example our variants can cover the space of code  ...  Introduction Computer systems rarely exploit the underlying hardware to its fullest potential.  ... 
doi:10.1007/978-3-642-00722-4_18 fatcat:qzezq2lxyjbwrndvroqf6dn33e

Pedestrian Detection at Warp Speed: Exceeding 500 Detections per Second

Floris De Smedt, Kristof Van Beeck, Tinne Tuytelaars, Toon Goedeme
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
Furthermore we present evaluation results on a very specific application showing the full potential of this warping window approach: detection of pedestrians in a truck's blind spot zone.  ...  We achieve these results by combining our fast pedestrian detection algorithm (implemented as a hybrid CPU and GPU combination) with the exploitation of scene constraints (using a warping window approach  ...  Multi-threaded hybrid implementation Although the use of GPU hardware allows for a significant speed-up, it does not fully exploit the capabilities of the hardware system, since the CPU is only active  ... 
doi:10.1109/cvprw.2013.94 dblp:conf/cvpr/SmedtBTG13 fatcat:mt3pqt54hfes7nc3bi46apgkt4

Forgery Detection and Value Identification of Euro Banknotes

Arcangelo Bruna, Giovanni Farinella, Giuseppe Guarnera, Sebastiano Battiato
2013 Sensors  
The hardware does not use any mechanical parts, so the overall system is low-cost. The proposed solution is reliable for ambient light and banknote positioning.  ...  This paper describes both hardware and software components to detect counterfeits of Euro banknotes. The proposed system is also able to recognize the banknote values.  ...  Acknowledgments The authors would like to thank Imperial Emporium Srl [19] and the Bank of Italy [5] for supporting this research activity.  ... 
doi:10.3390/s130202515 pmid:23429514 pmcid:PMC3649387 fatcat:fyovtx3mifflfiyiw52k5ydhem

Fpga-Oriented Secure Data Path Design: Implementation of a Public Key Coprocessor

Nele Mentens, Kazuo Sakiyama, Lejla Batina, Ingrid Verbauwhede, Bart Preneel
2006 2006 International Conference on Field Programmable Logic and Applications  
We overcome these drawbacks by fitting the public key architecture and algorithms into a coprocessor that optimally exploites the dedicated features on a Spartan XC3S4000.  ...  This feature does not only cause an inevitable performance degradation, but also an area increase.  ...  A hardware optimized version of Montgomery multiplication is used for modular multiplication. The so-called dual processor can operate in parallel for ECC or in a pipelined manner for RSA.  ... 
doi:10.1109/fpl.2006.311205 dblp:conf/fpl/MentensSBVP06 fatcat:fhw4szm75ranpit2mebsiawmmu

Syndrome-Enabled Unsupervised Learning for Channel Adaptive Blind Equalizer with Joint Optimization Mechanism [article]

Chieh-Fang Teng, Yen-Liang Chen
2020 arXiv   pre-print
Furthermore, the proposed syndrome-enabled blind equalizer can avoid the transmission of training sequences under time-varying fading channel and achieve global optimum via joint optimization mechanism  ...  , which has 1.3 dB gain over non-blind minimum mean square error (MMSE) equalizer.  ...  In [1] - [2] , convolutional neural networks are exploited for powerful modulation classification.  ... 
arXiv:2001.01426v1 fatcat:ny74gy3fcnfgnagzlysim7trxm

Energy-Efficient Processor for Blind Signal Classification in Cognitive Radio Networks

Eric Rebeiz, Fang-Li Yuan, Paulo Urriza, Dejan Markovic, Danijela Cabric
2014 IEEE Transactions on Circuits and Systems Part 1: Regular Papers  
However, when the signal parameters are unknown, an exhaustive search for cyclostationary features is energy inefficient due to high computational complexity.  ...  Second, we optimize the processor architecture by the co-design methodology to enhance block reusability and reconfigurability.  ...  We then exploit the functional similarities between algorithms to build a processing architecture that maximizes hardware utilization.  ... 
doi:10.1109/tcsi.2013.2278392 fatcat:bqfvl5q2e5gvzfi5hm3sjpl57a

Towards a Time-predictable Dual-Issue Microprocessor: The Patmos Approach

Martin Schoeberl, Pascal Schleuniger, Wolfgang Puffitsch, Florian Brandner, Christian W. Probst, Marc Herbstritt
2011 Design, Automation, and Test in Europe  
Many architectural features that increase the average case performance are hard to be modeled for the WCET analysis.  ...  Current processors are optimized for average case performance, often leading to a high worst-case execution time (WCET).  ...  WCET-aware Compilation The Patmos approach relies on a strong compiler in order to optimally exploit the available hardware resources.  ... 
doi:10.4230/oasics.ppes.2011.11 dblp:conf/date/SchoeberlSPBP11 fatcat:f3mbwaezuvbeppzfkclo426g2q

Blind identification strategies for room occupancy estimation

A. Ebadat, G. Bottegal, D. Varagnolo, B. Wahlberg, H. Hjalmarsson, K. H. Johansson
2015 2015 European Control Conference (ECC)  
The first tier is a blind identification step, based either on a frequentist Maximum Likelihood method, implemented using non-linear optimization, or on a Bayesian marginal likelihood method, implemented  ...  steps are blind.  ...  There are several approaches to address occupancy estimation: i) use dedicated hardware for counting people, [5] , [6] , [7] ; ii) use only information from non-dedicated hardware, already installed  ... 
doi:10.1109/ecc.2015.7330720 dblp:conf/eucc/EbadatBVWHJ15 fatcat:kx7wadpkrjahba42sqblh6s4ce

Blind spectrum sensing using symmetry property of cyclic autocorrelation function: from theory to practice

Lise Safatly, Babar Aziz, Amor Nafkha, Yves Louet, Youssef Nasser, Ali El-Hajj, Karim Y Kabalan
2014 EURASIP Journal on Wireless Communications and Networking  
This study shows that the blind cyclostationary feature detector outperforms the classical energy detector while guaranteeing acceptable complexity and low sensing time.  ...  In this paper, a blind cyclostationary feature detector, which is based on the symmetry property of cyclic autocorrelation function (SP-CAF), is implemented and tested using universal software radio peripheral  ...  It has been also supported by the Lebanese National Council for Scientific Research.  ... 
doi:10.1186/1687-1499-2014-26 fatcat:prvwdmatrbe5taqnjgfgitrlxy

2020 Index IEEE Transactions on Circuits and Systems for Video Technology Vol. 30

2020 IEEE transactions on circuits and systems for video technology (Print)  
., +, TCSVT Jan. 2020 232-242 Exploiting Trigonometric Properties to Optimize Higher Order DCT Architecture in HEVC.  ...  Buades, A., +, TCSVT July 2020 1960-1974 Exploiting Trigonometric Properties to Optimize Higher Order DCT Archi- tecture in HEVC.  ...  A Memory-Efficient Hardware Architecture for Connected Component Labeling in Embedded System.  ... 
doi:10.1109/tcsvt.2020.3043861 fatcat:s6z4wzp45vfflphgfcxh6x7npu

Serving DNNs in Real Time at Datacenter Scale with Project Brainwave

Eric Chung, Jeremy Fowers, Kalin Ovtcharov, Michael Papamichael, Adrian Caulfield, Todd Massengill, Ming Liu, Daniel Lo, Shlomi Alkalay, Michael Haselman, Maleen Abeydeera, Logan Adams (+31 others)
2018 IEEE Micro  
Exploiting distributed model parallelism and pinning over low-latency hardware microservices, Project Brainwave serves state-of-the-art, pre-trained DNN models with high efficiencies at low batch sizes  ...  Project Brainwave, Microsoft's principal infrastructure for AI serving in real time, accelerates deep neural network (DNN) inferencing in major services such as Bing's intelligent search features and Azure  ...  Project Brainwave, Microsoft's principal infrastructure for accelerated AI serving in the cloud, successfully exploits FPGAs on a datacenter-scale fabric for real-time serving of state-of-the-art DNNs.  ... 
doi:10.1109/mm.2018.022071131 fatcat:6cdzotc6bnb5pa2yw74n2o6xq4

Ensemble learning of diffractive optical networks [article]

Md Sadman Sakib Rahman, Jingxi Li, Deniz Mengu, Yair Rivenson, Aydogan Ozcan
2020 arXiv   pre-print
Specifically, there has been a revival of interest in optical computing hardware, due to its potential advantages for machine learning tasks in terms of parallelization, power efficiency and computation  ...  Here, we significantly improve the inference performance of diffractive optical networks using feature engineering and ensemble learning.  ...  For example, for the ensemble of D 2 NNs depicted in Fig. 3 , if a simple additive sum of the individual class scores is used instead of the optimized class-specific weights, the blind classification  ... 
arXiv:2009.06869v1 fatcat:4ldje5iyn5egzewl3bhulc6ik4

Guest Editorial Computational Imaging for Earth Sciences

Shuchin Aeron, Eric L. Miller, Melba Crawford, Alison Malcom, Andreas Reigber, Jocelyn Chanussot
2017 IEEE Transactions on Computational Imaging  
Fisher greatly for her patience and diligence in enabling proper and timely handling of the manuscripts.  ...  The editors would like to thank the reviewers who provided timely reviews and constructive suggestions for the papers. The editors appreciate the efforts of the IEEE publications co-ordinator A.  ...  A convex optimization algorithm is proposed for efficient estimation with significant gains over the sate-of-the-art.  ... 
doi:10.1109/tci.2017.2694978 fatcat:7jfk5xhrlner3forntsrzixeti

Ensemble learning of diffractive optical networks

Md Sadman Sakib Rahman, Jingxi Li, Deniz Mengu, Yair Rivenson, Aydogan Ozcan
2021 Light: Science & Applications  
Specifically, there has been a revival of interest in optical computing hardware due to its potential advantages for machine learning tasks in terms of parallelization, power efficiency and computation  ...  Here, we substantially improve the inference performance of diffractive optical networks using feature engineering and ensemble learning.  ...  neural networks, where we exploit the parallel processing of optical information.  ... 
doi:10.1038/s41377-020-00446-w pmid:33431804 fatcat:ij7yl7ebsnbwfowf3vzy623gky

Sponge Examples: Energy-Latency Attacks on Neural Networks [article]

Ilia Shumailov, Yiren Zhao, Daniel Bates, Nicolas Papernot, Robert Mullins, Ross Anderson
2021 arXiv   pre-print
We show how adversaries can exploit carefully crafted sponge examples, which are inputs designed to maximise energy consumption and latency.  ...  Our attacks can also be used to delay decisions where a network has critical real-time performance, such as in perception for autonomous vehicles.  ...  In this blind adversary setting, we exploit transferability across both models and hardware. Indeed, in Section 4.5 and Appendix F, we show that sponge examples transfer across models.  ... 
arXiv:2006.03463v2 fatcat:n7rgs3x3j5e7ljabfxdifpmfpe
« Previous Showing results 1 — 15 out of 14,500 results