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.
Dropout is a very effective way of regularizing neural networks. Stochastically “dropping out” units with a certain probability discourages over-specific co-adaptations of feature detectors, preventing overfitting and improving network generalization. Besides, Dropout can be interpreted as an approximate model aggregation technique, where an exponential number of smaller networks are averaged in order...
Training deep neural networks is difficult for the pathological curvature problem. Re-parameterization is an effective way to relieve the problem by learning the curvature approximately or constraining the solutions of weights with good properties for optimization. This paper proposes to reparameterize the input weight of each neuron in deep neural networks by normalizing it with zero-mean and unit-norm,...
In this work we propose a novel framework named Dual-Net aiming at learning more accurate representation for image recognition. Here two parallel neural networks are coordinated to learn complementary features and thus a wider network is constructed. Specifically, we logically divide an end-to-end deep convolutional neural network into two functional parts, i.e., feature extractor and image classifier...
Deep neural networks enjoy high interest and have become the state-of-art methods in many fields of machine learning recently. Still, there is no easy way for a choice of network architecture. However, the choice of architecture can significantly influence the network performance. This work is the first step towards an automatic architecture design. We propose a genetic algorithm for an optimization...
A pre-trained convolutional deep neural network (CNN) is widely used for embedded systems, which requires highly power-and-area efficiency. In that case, the CPU is too slow, the embedded GPU dissipates much power, and the ASIC cannot keep up with the rapidly progress of the CNN variations. This paper uses a binarized CNN which treats only binary 2-values for the inputs and the weights. Since the...
The development of a deep (stacked) convolutional auto-encoder in the Caffe deep learning framework is presented in this paper. We describe simple principles which we used to create this model in Caffe. The proposed model of convolutional auto-encoder does not have pooling/unpooling layers yet. The results of our experimental research show comparable accuracy of dimensionality reduction in comparison...
An architecture and learning methods for a growing neuro-fuzzy system that enlarges an amount of layers and tunes their synaptic weights in an online way are introduced in the paper. A structure of the hybrid system is built with the help of extended neo-fuzzy neurons which are characterized by improved approximating capabilities. The main peculiar feature of the introduced system is a learning method...
Deep learning obtains successful results in solving many machine learning problems. In this study, image classification process is performed by using Convolutional Neural Network (CNN) which is the most used architecture of deep learning. Image classification is used in a lot of basic field like medicine, education and security. Conditions that correct classification has vital importance may be especially...
In this paper, we discuss the architecture exploration of a Neuromorphic Signal Processing Integrated Circuit using Precise Timing. This device is intended to fulfill the role of a Digital Signal Processor in the spiking domain, becoming an essential tool to Spiking Neuromorphic Sensors such as Dynamic Vision Sensors. Our approach is based on the use of Spiking Neural Networks with preset topology...
Area V5 or Middle Temporal (MT) area of the primate brain is said to be involved in visual motion perception. Physiological studies indicate that the neurons in MT respond selectively to the direction of moving stimuli. However in response to the complex stimuli containing multiple oriented components, a set of MT neurons are selective to the direction of the component motion whereas the other set...
Automatic recognition of human demographical attributes has implications in a variety of domains, such as surveillance systems, human computer interaction, marketing etc. In this paper, we present an automatic gender recognition method from facial images based on convolutional neural networks. In order to train the network, we merged together several face databases and also gathered and annotated...
In general, the three main modules of the color scene classification systems are image decolorization, feature extraction and classification. The work presented in this paper focuses on image decolorization and classification as two stages. The first stage or objective of this paper is to improve the performance of the color scene classification system using deep belief networks (DBN) and support...
In this paper, we review different memristive threshold logic (MTL) circuits that are inspired from the synaptic action of the flow of neurotransmitters in the biological brain. The brainlike generalization ability and the area minimization of these threshold logic circuits aim toward crossing Moore’s law boundaries at device, circuits, and systems levels. Fast switching memory, signal processing,...
In this paper we present a memristive neuromorphic system for higher power and area efficiency. The system is based on a mixed signal approach considering the digital nature of the peripheral and control logics and the integration being analog. So, the system is connected digitally outside but the core is purely analog. This mixed signal approach provides the advantage of implementing neural networks...
Scaling down the transistor to gain more computation power will eventually reach the unsurmountable physical limitation. Neuromorphic Computing is a novel and promising computing scheme emulating the nervous structure and data processing methodology of a human brain. This paper presents the comparison between the conventional Von Neumann architecture and the neuromorphic computing architecture. We...
The expanding use of deep learning algorithms causes the demands for accelerating neural network (NN) signal processing. For the NN processing, in-memory computation is desired, in which expensive data transfer can be eliminated. In reflection of recently proposed binary neural networks (BNNs), which can reduce the computation resource and area requirements, we designed an in-memory BNN signal processor...
In this paper, a novel neural network architecture is proposed which results in an area-efficient feed-forward network. These structures require high-resolution multipliers. In order to overcome this problem, a mixed-signal Multiplying Digital to Analog Converter (MDAC) architecture which employs Delta-Sigma Modulation (DSM) to encode the multiplication results into the time domain. The time-domain...
Spiking Neural Networks (SNNs) are the third generation of artificial neural networks that closely mimic the time encoding and information processing aspects of the human brain. It has been postulated that these networks are more efficient for realizing cognitive computing systems compared to second generation networks that are widely used in machine learning algorithms today. In this paper, we review...
The paper presents the results of research on the use of Deep Neural Networks (DNN) for automatic classification of the skin lesions. The authors have focused on the most effective kind of DNNs for image processing, namely Convolutional Neural Networks (CNN). In particular, three kinds of CNN were analyzed: VGG19, Residual Networks (ResNet) and the hybrid of VGG19 CNN with the Support Vector Machine...
In this paper we propose a neural network allowing a mobile robot to learn artwork appreciation. The learning is based on the social referencing approach. The robot acquires its knowledge (artificial taste) from the interaction with humans. We present and analyze specifically the visual system, its impact on the robot behavior, and at the end, we analyze the readability of our robot behavior according...
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.