Year-end Sale % OFF 
Abstract
The paper demonstrates the advantages of the deep learning networks over the ordinary neural networks on their comparative applications to image classifying. An autoassociative neural network is used as a standalone autoencoder for prior extraction of the most informative features of the input data for neural networks to be compared further as classifiers. The main efforts to deal with deep learning networks are spent for a quite painstaking work of optimizing the structures of those networks and their components, as activation functions, weights, as well as the procedures of minimizing their loss function to improve their performances and speed up their learning time. It is also shown that the deep autoencoders develop the remarkable ability for denoising images after being specially trained. Convolutional Neural Networks are also used to solve a quite actual problem of protein genetics on the example of the durum wheat classification. Results of our comparative study demonstrate the undoubted advantage of the deep networks, as well as the denoising power of the autoencoders. In our work we use both GPU and cloud services to speed up the calculations.
Highlights
The physicists have accumulated a solid experience on the use of artificial neural networks (ANN) to the solution of many high energy and nuclear physics (HENP) problems [1]
We propose a deep learning model for the image classification, which consists of two neural networks: an autoencoder for preliminary input compression and an estimator for classifying images
A comparative study of three neural classifiers: a perceptron with one hidden layer, Deep Belief Networks (DBN) and Deep Neural Network (DNN) was accomplished on images of two famous benchmark data sets MNIST and FERET being previously compressed by the autoencoder
Summary
The physicists have accumulated a solid experience on the use of artificial neural networks (ANN) to the solution of many high energy and nuclear physics (HENP) problems [1]. An important role in this experience plays the remarkable privilege to generate training samples of any arbitrary length needed for sufficient ANN training by Monte Carlo simulations of advanced theoretical models Such comfortable circumstances largely distracted the physicists from the new deep learning neural network (NN) paradigm proposed in the newcoming challenging “Big Data” era asking for intensive data management. Hinton [2] as a combination of several stacked layers of recurrent NN of Boltzmann machine type realizing unsupervised representations of input data with the last supervised layer completing a classification task This type of deep networks was named Deep Belief Networks (DBN) referring one to the stochastic NNs and the probability distribution of their neurons. Being trained by a labeled sample with the back-prop method, CNN obtains all weights of convolutional filters and constructs a library of image elements linked to different image classes (see exhaustive CNN description in [7])
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
