A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
On the Expressive Power of Deep Neural Networks
[article]
2017
arXiv
pre-print
Preserved Fulltext
Other Versions
We propose a new approach to the problem of neural network expressivity, which seeks to characterize how structural properties of a neural network family affect the functions it is able to compute. ...
Our approach is based on an interrelated set of measures of expressivity, unified by the novel notion of trajectory length, which measures how the output of a network changes as the input sweeps along ...
The contributions of this paper are thus: (1) Measures of expressivity: We propose easily computable measures of neural network expressivity that capture the expressive power inherent in different neural ...
arXiv:1606.05336v6
fatcat:u64esvmisrfljkdgiymimdwlhe
Understanding and Training Deep Diagonal Circulant Neural Networks
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we study deep diagonal circulant neural networks, that is deep neural networks in which weight matrices are the product of diagonal and circulant ones. ...
We conduct a thorough experimental study to compare the performance of deep diagonal circulant networks with state of the art models based on structured matrices and with dense models. ...
Unfortunately, this theorem is of little use to understand the expressive power of diagonal-circulant matrices when they are used in deep neural networks. ...
arXiv:1901.10255v3
fatcat:c3gbjaczqrblpdbiji7g4qyom4
Deep neural nets based power amplifier non-linear pre-distortion
2017
Journal of Physics, Conference Series
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
This paper proposed a novel method based on deep neural networks (auto-encoder) model, to construct the pre-distortion model for non-linear feature of power amplifier. ...
The experimental results show the effectiveness and efficiency of deep neural network based power amplifier non-linear predistortion technique. ...
The Pre-distortion Model Based on Deep Neural Network This section introduces the role of the deep neural network in the pre-distortion model and concrete implementation methods. ...
doi:10.1088/1742-6596/887/1/012049
fatcat:bjtgwqof5ze3xhmggj6szmocpm
Time for a Background Check! Uncovering the impact of Background Features on Deep Neural Networks
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
With increasing expressive power, deep neural networks have significantly improved the state-of-the-art on image classification datasets, such as ImageNet. ...
In this paper, we investigate to what extent the increasing performance of deep neural networks is impacted by background features? ...
Introduction One key driver behind this success of modern deep neural networks (DNNs) is their expressive power, which enables them to learn a rich set of representations required to solve a target task ...
arXiv:2006.14077v1
fatcat:kfbxsnwqyvg3dgexuyqobd5hnm
Physics-Enforced Modeling for Insertion Loss of Transmission Lines by Deep Neural Networks
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2107.12527v1 fatcat:z3blppekcvfo7b2alljmhk34i4
The resulting neural network is applied to predict the coefficients of the polynomial expression. ...
We first show that direct application of neural networks can lead to non-physics models with negative insertion loss. To mitigate this problem, we propose two deep learning solutions. ...
Then, we apply the neural networks to find the coefficients of the expression. The new architecture of neural networks is shown in Fig. 3 . dimensional parameters and coefficients, respectively. ...
arXiv:2107.12527v2
fatcat:fqie2h4xizbdvgeo36vqprtlba
Optimizing Deep Convolutional Neural Network for Facial Expression Recognition
2020
European Journal of Engineering Research and Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We designed a large, deep convolutional neural network to classify 40,000 images in the data-set into one of seven categories (disgust, fear, happy, angry, sad, neutral, surprise). ...
In this project, we have designed deep learning Convolution Neural Network (CNN) for facial expression recognition and developed model in Theano and Caffe for training process. ...
ACKNOWLEDGMENT The authors would like to thank NVIDIA corporation for donating NVIDIA GPU card for this research work. ...
doi:10.24018/ejers.2020.5.2.495
fatcat:euv4bc7cvjcqnaevup43xjzbaq
Model Complexity of Deep Learning: A Survey
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Model complexity is a fundamental problem in deep learning. In this paper we conduct a systematic overview of the latest studies on model complexity in deep learning. ...
Model complexity of deep learning can be categorized into expressive capacity and effective model complexity. ...
[92] find that smaller widths in the first few layers of a deep ReLU network cause a bottleneck on the expressive power. Kileel et al. ...
arXiv:2103.05127v2
fatcat:uknedzqea5evdcqm7mnlatytya
EmotionNet Nano: An Efficient Deep Convolutional Neural Network Design for Real-time Facial Expression Recognition
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Experimental results using the CK+ facial expression benchmark dataset demonstrate that the proposed EmotionNet Nano networks demonstrated accuracies comparable to state-of-the-art in FEC networks, while ...
neural network created through a human-machine collaborative design strategy, where human experience is combined with machine meticulousness and speed in order to craft a deep neural network design catered ...
complexities of the deep neural networks that drive such systems. ...
arXiv:2006.15759v1
fatcat:kjxpea3sgngexlgpleoukvhrfa
Robot formation control based on Internet of things technology platform
2020
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Finally, the simulation of robot formation motion is established by MATLAB software, which verifies the feasibility of particle swarm optimization deep learning neural network algorithm under the Internet ...
In order to meet the efficient response requirements of robot formation control, a real-time transmission system of robot cooperative motion control is built based on the Internet of things platform, which ...
FIGURE 8 . 8 Training process of deep learning neural network based on particle swarm optimization.
FIGURE 9 . 9 Performance evolution of deep learning neural network. ...
doi:10.1109/access.2020.2992701
fatcat:7wobxm66xnbx5kk2mcqct46gby
EmotionNet Nano: An Efficient Deep Convolutional Neural Network Design for Real-Time Facial Expression Recognition
2021
Frontiers in Artificial Intelligence
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
To the best of the author's knowledge, this is the very first deep neural network architecture for facial expression recognition leveraging machine-driven design exploration in its design process, and ...
Experimental results using the CK + facial expression benchmark dataset demonstrate that the proposed EmotionNet Nano networks achieved accuracy comparable to state-of-the-art FEC networks, while requiring ...
assembled and organized the database. JL and LW ran the experiments and performed statistical analysis. JL wrote the first draft of the manuscript. ...
doi:10.3389/frai.2020.609673
pmid:33733225
pmcid:PMC7861268
fatcat:ckzikei5hreejespczhim4guo4
Survey of Expressivity in Deep Neural Networks
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We survey results on neural network expressivity described in "On the Expressive Power of Deep Neural Networks". ...
They suggest that parameters earlier in a network have greater influence on its expressive power -- in particular, given a layer, its influence on expressivity is determined by the remaining depth of the ...
Motivation and Setting In this survey, we summarize results on the expressivity of deep neural networks from [1] . ...
arXiv:1611.08083v1
fatcat:wbzrm6x3rrf7vlknn5fekqlib4
GradNets: Dynamic Interpolation Between Neural Architectures
[article]
2015
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Neural Networks, in particular, have enormous expressive power and yet are notoriously challenging to train. The nature of that optimization challenge changes over the course of learning. ...
Traditionally in deep learning, one makes a static trade-off between the needs of early and late optimization. ...
ACKNOWLEDGMENTS We thank NVIDIA for their generosity in providing access to part of their cluster in support of Enlitic's mission and our research. ...
arXiv:1511.06827v1
fatcat:kkpfwhtlt5anlkc3zitevgcxue
Recognition of Emotion From Facial Expression for Autism Disorder
2019
International journal of recent technology and engineering
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
From the point of view of automatic recognition, The facial expression may included the figurations of the facial parts and their spatial relationships or changes in the pigmentation of the face. ...
Use The Camera to capture the live images of autism people ...
data set and perform deep neural network we classify the dataset into different emotions and with the help of the classification we train the model using deep convolutional neural network where the data ...
doi:10.35940/ijrte.b1095.0782s319
fatcat:6sqqyhgkznbexms6brz6lf454i
Predicting Power Density of Array Antenna in mmWave Applications with Deep Learning
2021
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In Table III , N1 and N2 are three layers networks that consists of one input layer, one hidden layer, and one output layer, and the number of nodes inside each layer is expressed as L1, L2, and L3, respectively ...
To improve the efficiency of this system, various studies that are based on neural networks have been conducted. ...
doi:10.1109/access.2021.3102825
fatcat:6svfqo7em5b3ncnoeuis6ve6o4
Exploiting more robust and efficacious deep learning techniques for modeling wind power with speed
2021
Energy Reports
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Sound analyses of the nonlinear relationship between wind speed and power generation are crucial for the advancement of wind energy optimization. ...
This paper assesses and presents a number of model-control techniques, categorized as model-oriented and data-oriented, to achieve more robust and efficacious deep neural networks for applications in the ...
Simple three-layer neural networks and deep multilayer networks can assist in basic simulations of the nonlinear relationship between wind speed and power. ...
doi:10.1016/j.egyr.2021.11.151
fatcat:t5ujsxfueffjfpvaij6u5x66ke
« Previous
Showing results 1 — 15 out of 133,386 results