The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Although Deep Convolutional Neural Networks (CNNs) have liberated their power in various computer vision tasks, the most important components of CNN, convolutional layers and fully connected layers, are still limited to linear transformations. In this paper, we propose a novel Factorized Bilinear (FB) layer to model the pairwise feature interactions by considering the quadratic terms in the transformations...
Ship detection is a fundamental task for SAR-based maritime surveillance. Besides providing high reliability, a good detector is required to be computationally light, in order to analyze huge areas in a reasonable time. We propose a fully convolutional neural network for ship detection in SAR images. Thanks to a relatively simple architecture, complexity remains low enough to allow for a single-stage...
Machine Learning (ML) is an attractive application of Non-Volatile Memory (NVM) arrays [1,2]. However, achieving speedup over GPUs will require minimal neuron circuit sharing and thus highly area-efficient peripheral circuitry, so that ML reads and writes are massively parallel and time-multiplexing is minimized [2]. This means that neuron hardware offering full ‘software-equivalent’ functionality...
Extreme learning machine is an emerging neural network architecture that offers fast learning and generalization for multiple tasks. In this work, a scalable digital architecture for multi-classifier extreme learning machine (MT-ELM) is proposed. The proposed architecture performs multiple classification tasks without reconfiguring the network. The design is validated with MNIST dataset and it is...
This work attempts to find the most optimal setting for shallow artificial neural network (ANN) for Bengali digit dataset. Recognition of handwritten Bengali numerals has recently gained much interest among researchers due to significant performance gain found in the recognition of English numerals using artificial neural network. In this work, a new dataset of 70,000 samples were created first by...
A conventional weight in an artificial neural network has a single trainable real value and produces a linear relationship between the weight input and the weight output. A real synaptic cleft is also trainable but provides a more complex relationship. It is obvious to wonder if adding extra complexity to the conventional weight response would lead to more capable networks. This work describes a weight...
Echo State Network (ESN) is a type of neural network with convex training used as nonlinear adaptive filters. It requires relative large network size and O(N2) training algorithm in order to realize its full potential. We present a subspace training scheme for ESN that reduces the training cost from O(N2) to O(NL) where L < N is the dimension of the subspace. Experiments show that L = 20 is sufficient...
In supervised learning, one of the motivations and intuitions behind the design of the learning algorithm is to prevent an overfitting. Overfitting usually occurs when a learning model is excessively complex, especially when having too many parameters to adjust. A typical approach to alleviate this problem is to introduce a form of the penalty term in the objective function, which is called regularization,...
This paper presents an approach to build a data classifier based on a simple and inexpensive evaluation function aimed to reduce the computational costs when processing new incoming instances. The classifier agent employs in its training concepts of Self-Organized Maps and Multiple Instance Learning. The motivation for this proposal was the need of a classifier for the processing of signals from partial...
Determination of model complexity is a challenging issue to solve computer vision problems using restricted boltzmann machines (RBMs). Many algorithms for feature learning depend on cross-validation or empirical methods to optimize the number of features. In this work, we propose an learning algorithm to find the optimal model complexity for the RBMs by incrementing the hidden layer. The proposed...
This paper presents an approach to build a data classifier based on a simple and inexpensive evaluation function aimed to reduce the computational costs when processing new incoming instances. The classifier agent employs concepts of Self-Organized Maps and Multiple Instance Learning. The motivation for this proposal was the need of a classifier for the processing of signals from partial discharges...
Though the Extreme Learning Machine (ELM) has become quite popular in recent years, there are no performance guarantees; the resultant networks also tend to be densely connected. The complexity of a learning machine may be measured by the Vapnik-Chervonenkis (VC) dimension, and a small VC dimension leads to good generalization and lower test set errors. The Minimal Complexity Machine (MCM), that has...
In this article the approximation capability of the extreme learning machine is studied. Specifically the impact of the range from which the input weights and biases are randomly generated on the fitted curve complexity is analyzed. The guidance for how to generate the input weights and biases to get good performance in approximation of the functions of one variable is provided.
A non-parametric probability density function (pdf) estimation technique is presented. The estimation consists in approximating the unknown pdf by a network of Gaussian Radial Basis Functions (GRBFs). Complexity analysis is introduced in order to select the optimal number of GRBFs. Results obtained on real data show the potentiality of this technique.
Neural Network making use of Radial Basis Function (RBF) in the hidden layer maps the input of a lower dimension to a higher dimensional space in order to make the input linearly separable. The traditional RBF model is normally referred as cognitive component. The major issues in the traditional model are large number of fixed neurons, use of complete training set, prior center selection etc,. These...
Recently, the application of complex-valued neural networks (CVNNs) for real-valued classification has attracted more and more attention. To overcome the limitations of the existing CVNNs, this study extends the real-valued group method of data handling (RGMDH) type neural network to complex domain, and constructs complex-valued GMDH-type neural network (CGMDH). First, it proposes the complex least...
One of the disadvantages of using Artificial Neural Networks (ANNs) is their significant training time need, which scales with the complexity of the network and with the complexity of the problem that is needed to be solved. Radial Basis Function Neural Networks (RBFNNs) are neural networks that use the linear combination of radial basis functions, utilizing hybrid learning procedures which can solve...
Parallel implementation of neural networks is amongst major area of research in computer science. Self Organizing Map (SOM) is a neural network that has been under spotlight throughout last decade for implementation in parallel architecture. SOM trains itself through unsupervised learning by retrieving inherent topological features of applied input data. In this paper design and implementation of...
Aimed at the problem of the automatic detection and classification identification for granary pests, an automatic granary pest identification system based on morphological feature is designed. Five morphological features such as complexity, duty cycle, elongation, moment invariants 1 and 2 are automatically extracted from granary pest image, and loaded into a BP neural network as input factors. The...
Software cost estimation is a crucial element in project management. Failing to use a proper cost estimation method might lead to project failures. According to the Standish Chaos Report, 65% of software projects are delivered over budget or after the delivery deadline. Conducting software cost estimation in the early stages of the software life cycle is important and this would be helpful to project...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.