Filters








225 Hits in 8.3 sec

Deep Neural Network Approximation for Custom Hardware: Where We've Been, Where We're Going [article]

Erwei Wang, James J. Davis, Ruizhe Zhao, Ho-Cheung Ng, Xinyu Niu, Wayne Luk, Peter Y. K. Cheung, George A. Constantinides
2019 arXiv   pre-print
This article represents the first survey providing detailed comparisons of custom hardware accelerators featuring approximation for both convolutional and recurrent neural networks, through which we hope  ...  Deep neural networks have proven to be particularly effective in visual and audio recognition tasks.  ...  Ba et al. combined low-rank factorisation with knowledge distillation, where a deep and complex neural network is mimicked with a simpler, shallower one [11] .  ... 
arXiv:1901.06955v3 fatcat:rkgo2oisdrgv3dtnbtlldlkpba

Hardware Developments Iii

Alan Ó Cais, Liang Liang, Jony Castagna
2018 Zenodo  
and detailed feedback to the project software developers; - discussion of project software needs with hardware and software vendors, completion of survey of what is already available for particular hardware  ...  Update on "Hardware Developments II" (Deliverable 7.3: https://doi.org/10.5281/zenodo.1207613) which covers: - Report on hardware developments that will affect the scientific areas of interest to E-CAM  ...  In particular, deep learning needs a very fast connection between GPUs to allow fast training of Neural Networks (NN) on multi-GPUs.  ... 
doi:10.5281/zenodo.1304087 fatcat:itkihkoikvas5ajgxzqyswsez4

Hardware Developments Ii

Liang Liang, Jony Castagna, Alan O'Cais, Simon Wong, Goar Sanchez
2017 Zenodo  
detailed feedback to the project software developers; - discussion of project software needs with hardware and software vendors, completion of survey of what is already available for particular hardware  ...  Update on "Hardware Developments I" (Deliverable 7.1: https://doi.org/10.5281/zenodo.929533) which covers: - Report on hardware developments that will affect the scientific areas of interest to E-CAM and  ...  Today's advanced deep neural networks use algorithms, big data, and the computational power of the GPU (and other technologies) to change this dynamic.  ... 
doi:10.5281/zenodo.1207612 fatcat:p75hwqe5jjantcugbqrov7ryla

D7.9: Hardware developments V

Alan O'Cais, Christopher Werner, Simon Wong, Padraig Ó Conbhuí, Jony Castagna, Godehard Sutmann
2020 Zenodo  
and detailed feedback to the project software developers; - discussion of project software needs with hardware and software vendors, completion of survey of what is already available for particular hardware  ...  Update on "Hardware Developments IV" (Deliverable 7.7: https://doi.org/10.5281/zenodo.3256137) which covers: - Report on hardware developments that will affect the scientific areas of interest to E-CAM  ...  The third trend arises from Deep Neural Networks (DNN) for back propagation learning of complex patterns.  ... 
doi:10.5281/zenodo.3931510 fatcat:kelxr6ap5vfunpjisfpwiauwue

D7.7: Hardware developments IV

Alan Ó Cais, Jony Castagna, Godehard Sutmann
2019 Zenodo  
and detailed feedback to the project software developers; - discussion of project software needs with hardware and software vendors, completion of survey of what is already available for particular hardware  ...  Update on "Hardware Developments III" (Deliverable 7.5: https://doi.org/10.5281/zenodo.1304088) which covers: - Report on hardware developments that will affect the scientific areas of interest to E-CAM  ...  Today's advanced deep neural networks use algorithms, big data, and the computational power of the GPU (and other technologies) to change this dynamic.  ... 
doi:10.5281/zenodo.3256136 fatcat:hfpwvelb3zdxlk6fmkgddqgqoq

Autonomous Systems and the Challenges in Verification, Validation, and Test

David Yeh
2018 IEEE design & test  
As we've learned at this conference, these new systems depend on unsupervised machine learning.  ...  For various reasons, we don't have that data, and dealing with smaller data is a problem that we need to deal with because cognition is learning with smaller data more than learning with larger data.  ...  The neural network and natural languages efforts have been so successful that the potential applications are many. For example, the next thing we're going to see is the personal assistant.  ... 
doi:10.1109/mdat.2018.2816940 fatcat:jeoebmm665aozh752eydzatdmq

Introduction to Neural Network Verification

Aws Albarghouthi
2021 Foundations and Trends® in Programming Languages  
Deep learning has transformed the way we think of software and what it can do. But deep neural networks are fragile and their behaviors are often surprising.  ...  This monograph covers foundational ideas from formal verification and their adaptation to reasoning about neural networks and deep learning.  ...  Over the past decade, a number of hardware and software advances have conspired to thrust deep learning and neural networks to the forefront of computing.  ... 
doi:10.1561/2500000051 fatcat:2wbm374jcrc3pd5mhn5pfukoae

Introduction to Neural Network Verification [article]

Aws Albarghouthi
2021 arXiv   pre-print
Deep learning has transformed the way we think of software and what it can do. But deep neural networks are fragile and their behaviors are often surprising.  ...  This book covers foundational ideas from formal verification and their adaptation to reasoning about neural networks and deep learning.  ...  Is verification out of the question for deep neural networks? No!  ... 
arXiv:2109.10317v2 fatcat:abc6pneupzbrre2uwiamvnqk2e

Physics-based Deep Learning [article]

Nils Thuerey and Philipp Holl and Maximilian Mueller and Patrick Schnell and Felix Trost and Kiwon Um
2022 arXiv   pre-print
This digital book contains a practical and comprehensive introduction of everything related to deep learning in the context of physical simulations.  ...  RL has been very successful at playing games such as Go [SSS+17] , and it bears promise for engineering applications such as robotics.  ...  This is where so-called Bayesian neural network (BNN) approaches come into play.  ... 
arXiv:2109.05237v3 fatcat:pz7ot63dlbdkriihkwloefk3im

Computational Antitrust First Annual Conference: Exploring Antitrust 3.0

Thibault Schrepel
2022 Stanford Journal of Computational Antitrust  
It has been cleaned but not edited. Please use it accordingly and refer to the videos published on our YouTube channel.  ...  On December 13, 14, and 15, 2021, The Stanford Center for Legal Informatics (CodeX) organized the world's first (online) conference dedicated to Computational Antitrust under Prof.  ...  We then train the neural network on the documents for our cluster.  ... 
doi:10.51868/9 fatcat:ihmhoqcmsfhwfjdtucnr5pc6hu

Using Convolutional Neural Network for Image Classification and Segmentation

2022 Computer Engineering and Intelligent Systems  
In this paper, a deep neural network primarily based on Keras and TensorFlow is deployed using python.  ...  So, this paper introduces the idea of using different datasets and models of the deep learning network and comprehensively utilizes it to determine the best test accuracy for the images.  ...  As a deep neural network, CNN can capture complex features from image data.  ... 
doi:10.7176/ceis/13-1-03 fatcat:j3l2rwonuzcp3jokl42777emem

Transdisciplinary consolidation of financial technologies with knowledge systems
Трансдисциплінарна консолідація фінансових технологій із системами знань

Oleksandr LYUBICH, SESE "The Academy of Financial Management", Oleksandr STRYZHAK, SESE "The Academy of Financial Management"
2021 Fìnansi Ukraïni  
For applications where training data is scarce, neural networks can generate it.  ...  However, deep neural networks are woefully inefficient learners, requiring millions of images to learn how to detect objects.  ...  "It did require a paradigm shift for traditional aviators to recognize the value of a UAV," he said, "and I think the same thing has been true for operators of the surface fleet … [autonomous] submarine-hunting  ... 
doi:10.33763/finukr2021.11.054 fatcat:edc3aarta5gq7be2wixkynjz7q

Agents in industry: the best from the AAMAS 2005 industry track

M. Pechoucek, S.G. Thompson, J.W. Baxter, G.S. Horn, K. Kok, C. Warmer, R. Kamphuis, V. Maric, P. Vrba, K.H. Hall, F.P. Maturana, K. Dorer (+1 others)
2006 IEEE Intelligent Systems  
The articles in this department intend to give some indication of agent technology's readiness for commercial deployment, based primarily on the presentations and discussions at the inaugural Industry  ...  We're part of a QinetiQ team that's developing and implementing the decision-making partnership concept; we focus on the multiagentsystem element.  ...  Acknowledgments This research was part of the UK Ministry of Defence Output 3 research program on behalf of the Director Equipment Capability-Deep Target Attack.  ... 
doi:10.1109/mis.2006.19 fatcat:66r27spkmnacppwhzpu6rmco6e

Machine Learning, Social Learning and the Governance of Self-Driving Cars

Jack Stilgoe
2017 Social Science Research Network  
Selfdriving cars represent a high-stakes test of the powers of machine learning, as well as a test case for social learning in technology governance.  ...  As of 2016, these neural networks' greatest feats have been in digital image recognition, where it is claimed their abilities have surpassed humans' (Krizhevsky et al., 2012) , and voice recognition  ...  Following revelations about the potential of their GPU chips for use in deep neural networks in the late 2000s, the company has grown its machine learning business.  ... 
doi:10.2139/ssrn.2937316 fatcat:ph57apojq5fmzn4vy4gqh5ouni

Machine learning, social learning and the governance of self-driving cars

Jack Stilgoe
2017 Social Studies of Science  
Selfdriving cars represent a high-stakes test of the powers of machine learning, as well as a test case for social learning in technology governance.  ...  As of 2016, these neural networks' greatest feats have been in digital image recognition, where it is claimed their abilities have surpassed humans' (Krizhevsky et al., 2012) , and voice recognition  ...  Following revelations about the potential of their GPU chips for use in deep neural networks in the late 2000s, the company has grown its machine learning business.  ... 
doi:10.1177/0306312717741687 pmid:29160165 fatcat:r6ygqlloubcg5huznycahm2jnm
« Previous Showing results 1 — 15 out of 225 results