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.
Video scene parsing is challenging due to the following two reasons: firstly, it is non-trivial to learn meaningful video representations for producing the temporally consistent labeling map; secondly, such a learning process becomes more difficult with insufficient labeled video training data. In this work, we propose a unified framework to address the above two problems, which is to our knowledge...
Manual annotations of temporal bounds for object interactions (i.e. start and end times) are typical training input to recognition, localization and detection algorithms. For three publicly available egocentric datasets, we uncover inconsistencies in ground truth temporal bounds within and across annotators and datasets. We systematically assess the robustness of state-of-the-art approaches to changes...
Subspace learning is one of the most foundational tasks in computer vision with applications ranging from dimensionality reduction to data denoising. As geometric objects, subspaces have also been successfully used for efficiently representing certain types of invariant data. However, methods for subspace learning from subspace-valued data have been notably absent due to incompatibilities with standard...
Automatic grading systems, such as WebWork, are becoming much more widely used as they relieve the instructor from needing to grade student work, provide students with automatic feedback, and can allow for immediate resubmission. They have also been shown to improve the effectiveness of teaching and learning. In this paper, we apply Item Response Theory (IRT) to a large WebWork Calculus homework dataset...
Dropout is a technique widely used for preventing overfitting while training deep neural networks. However, applying dropout to a neural network typically increases the training time. This paper proposes a different dropout approach called controlled dropout that improves training speed by dropping units in a column-wise or row-wise manner on the matrices. In controlled dropout, a network is trained...
A hybrid sampling technique is proposed by combining Complementary Fuzzy Support Vector Machine (CMTFSVM) and Synthetic Minority Oversampling Technique (SMOTE) for handling the imbalanced classification problem. The proposed technique uses an optimised membership function to enhance the classification performance and it is compared with three different classifiers. The experiments consisted of four...
This study is focused on the application of the forecast results obtained with the help of a dual-parametric neural network for measuring instruments calibration within a limited range. In this report describing our study we intend to give a theoretical overview of the previous research on the subject. We also consider the learning algorithms for interval neural networks with hidden layer weights...
Online education constitutes an important component in the continuous growth / development of higher education and adult education. Its quality largely depends on the design rules and the evaluation standards of the programs as a whole as well as the study courses as component parts. The need to ensure the quality of online courses is a current issue. The paper proposes a system of standards, criteria...
In this paper, the problem of adaptive beamforming in the presence of direction-of-arrival (DOA) mismatch is investigated. To develop a robust beamformer against such an imperfection, a new approach is devised by formulating an output signal-to-interference-plus-noise ratio (SINR) maximization problem. In particular, the proposed robust beamforming approach consists of two steps. At first, the standard...
A method for scene text localization and recognition is proposed. The novelties include: training of both text detection and recognition in a single end-to-end pass, the structure of the recognition CNN and the geometry of its input layer that preserves the aspect of the text and adapts its resolution to the data.,,The proposed method achieves state-of-the-art accuracy in the end-to-end text recognition...
A power transformer fault diagnosis method based on Improved Particle Swarm Optimization and BP neural network is proposed. The particle swarm algorithm that used to optimize the parameters of the BP neural network is prone to “premature”. By optimizing the inertia weight, in the process of increasing the number of iterations, the inertia weight can be gradually reduced, and the algorithm can avoid...
Many existing scene parsing methods adopt Convolutional Neural Networks with fixed-size receptive fields, which frequently result in inconsistent predictions of large objects and invisibility of small objects. To tackle this issue, we propose a scale-adaptive convolution to acquire flexiblesize receptive fields during scene parsing. Through adding a new scale regression layer, we can dynamically infer...
We describe a method to produce a network where current methods such as DeepFool have great difficulty producing adversarial samples. Our construction suggests some insights into how deep networks work. We provide a reasonable analyses that our construction is difficult to defeat, and show experimentally that our method is hard to defeat with both Type I and Type II attacks using several standard...
When the training and the test data belong to different domains, the accuracy of an object classifier is significantly reduced. Therefore, several algorithms have been proposed in the last years to diminish the so called domain shift between datasets. However, all available evaluation protocols for domain adaptation describe a closed set recognition task, where both domains, namely source and target,...
Deep learning has greatly improved visual recognition in recent years. However, recent research has shown that there exist many adversarial examples that can negatively impact the performance of such an architecture. This paper focuses on detecting those adversarial examples by analyzing whether they come from the same distribution as the normal examples. Instead of directly training a deep neural...
Estimating depth from a single RGB image is an ill-posed and inherently ambiguous problem. State-of-the-art deep learning methods can now estimate accurate 2D depth maps, but when the maps are projected into 3D, they lack local detail and are often highly distorted. We propose a fast-to-train two-streamed CNN that predicts depth and depth gradients, which are then fused together into an accurate and...
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,...
Impressive image captioning results are achieved in domains with plenty of training image and sentence pairs (e.g., MSCOCO). However, transferring to a target domain with significant domain shifts but no paired training data (referred to as cross-domain image captioning) remains largely unexplored. We propose a novel adversarial training procedure to leverage unpaired data in the target domain. Two...
This work presents a method of classification of text documents using deep neural network with LSTM (long short-term memory) units. We have tested different approaches to build feature vectors, which represent documents to be classified: we used feature vectors constructed as sequences of words included in the documents, or, alternatively, we first converted words into vector representations using...
The basic features of some of the most versatile and popular open source frameworks for machine learning (TensorFlow, Deep Learning4j, and H2O) are considered and compared. Their comparative analysis was performed and conclusions were made as to the advantages and disadvantages of these platforms. The performance tests for the de facto standard MNIST data set were carried out on H2O framework for...
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.