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.
Sparsity helps reducing the computation complexity of DNNs by skipping the multiplication with zeros. The granularity of sparsity affects the efficiency of hardware architecture and the prediction accuracy. In this paper we quantitatively measure the accuracy-sparsity relationship with different granularity. Coarse-grained sparsity brings more regular sparsity pattern, making it easier for hardware...
This paper proposes a new intrinsic image decomposition method that decomposes a single RGB-D image into reflectance and shading components. We observe and verify that, a shading image mainly contains smooth regions separated by curves, and its gradient distribution is sparse. We therefore use ℓ1-norm to model the direct irradiance component — the main sub-component extracted from shading component...
Impaired cerebral autoregulation is often the cause of cerebral hemorrhage in preterm infants. The place of cerebral bleeding is most frequently the germinal matrix being a highly vascularized layer of neuronal and glial precursors. A passive, linear, dependency of the cerebral blood flow on the arterial mean pressure leads to damage of fragile blood vessels of the germinal matrix. A mathematical...
The paper considers the class of discrete-time, single-input, single-output, nonlinear dynamical systems described by a polynomial difference equation. This class, call polynomial time-invariant, is a proper generalization of the linear time-invariant model class. The identification data is assumed to be generated in the errors-in-variables setting, where the input and the output noise is zero mean,...
This paper introduces the development of a distributed air-defense engagement simulation model based on data distribution service (DDS). To design and develop effectively, system developers need a high-resolution engagement simulation including complex engineering-level models and operational scenario models. Increasing the resolution of the model results in the growing model's complexity which requires...
Bot Assistants can be an efficient and low-cost solution to Patient Care. One important aspect of Assistant Bots is successful Communication and Socialization with the patient. A new Conditional Entropy Retrieval Based model is proposed and also an Attitude Modeling based on Popitz Powers. The algorithm successfully retrieves the suitable answer with a high success rate in the patient-Bot Assistant...
The remaining useful life (RUL) prediction of bearings has emerged as a critical technique for providing failure warnings in advance, reducing costly unscheduled maintenance and enhancing the reliability of bearings. Recently, a fusion prognostics method combining exponential model and relevance vector machine (RVM) has been proposed and applied to the RUL prediction of bearings. This fusion prognostics...
In order to defend adversarial attacks in computer vision models, the conventional approach arises on actively incorporate such samples into the training datasets. Nonetheless, the manual production of adversarial samples is painful and labor intensive. Here we propose a novel generative model: Convolutional Autoencoder Model to add unsupervised adversarial training, i.e., the production of adversarial...
Optimum and heuristic sampling methods of a range of values of a two-dimensional random value are investigated. Conditions of their competence at recovery of the normal distribution law of two independent random values are defined.
This paper proposes a novel surrogate-model-based multi-objective evolutionary algorithm, which is called Multi-objective Bayesian Optimization Algorithm based on Decomposition (MOBO/D). In this algorithm, a multi-objective problem is decomposed into several subproblems which will be solved simultaneously. MOBO/D builds Gaussian process model for each objective to learn the optimization surface, and...
The authors propose a variational level set image segmentation method for intensity inhomogeneous texture image. The method first extracts the main image structure by a relative total variation image decomposition method, which can better decompose the image into structural and textural parts. Then only uses the structural part as the input image for the variational level set segmentation. The intensity...
This paper presents a budgetary learning algorithm for online multiclass classification. Based on the multiclass passive-aggressive learning with kernels, we introduce a dual perspective that gives rise to the proposed budgetary algorithm. Basically, the proposed algorithm limits the amount of data in use and fully exploits the available data on hand through optimization. The algorithm has both constant...
Radial basis function neural network has a strong capability of non-linear mapping for system identification. Especially, using the orthogonal least square method can generate a parsimonious structure to avoid “overfitting” problem effectively. Nonetheless, it is difficult to deal with dynamic systems by static models, which exist mainly in manufacture and life. Aimed at the non-stationary time series...
Due to increasing demand of low power computing, and diminishing returns from technology scaling, industry and academia are turning with renewed interest toward energy-efficient programmable accelerators. This paper proposes an Integrated Programmable-Array accelerator (IPA) architecture based on an innovative execution model, targeted to accelerate both data and control-flow parts of deeply embedded...
In this paper, we address the problem of nonlinear fault detection of chemical processes. The objective is to extend our previous work [1] to provide a better performance in terms of fault detection accuracies by developing a pre-image kernel PCA (KPCA)-based Generalized Likelihood Ratio Test (GLRT) technique. The benefit of the pre-image kPCA technique lies in its ability to compute the residual...
Nowadays, high-bandwidth networks are easily accessible in data centers. However, existing distributed graph-processing frameworks fail to efficiently utilize the additional bandwidth capacity in these networks for higher performance, due to their inefficient computation and communication models, leading to very long waiting times experienced by users for the graph-computing results. The root cause...
The problem of preference functions model development for multiple criteria decision-making is considered based on machine-learning approach. It is assumed that the training sample for a plurality of objects, for which decisions are made, is formed from a set of measured features or the particular criteria and the matrix of pairwise comparisons. The problem of constructing a linear preference function...
The paper is concerned with an increase in the accuracy of the dynamic input-output systems modeling with Volterra polynomials owing to a fuller consideration of data on system outputs to the test inputs. The methodology applied to construct the non-stationary Volterra polynomials is based on a priori consideration of necessary conditions for solvability of special multi-dimensional integral Volterra...
The objective of this paper is to extend the applicability of the GLR method to a wide range of practical systems. Most real systems are nonlinear, multivariate, and are best represented by input-output type of models. Kernel partial least squares (KPLS) models have been widely used to represent such systems. Therefore, in this paper, kernel PLS-based GLR method will be utilized in practice to improve...
How the brain maintains the stability of visual perception across saccade is a central question in systems neuroscience; accurately characterizing visual responses in the perisaccadic period is an important step towards understanding how the visual world is represented during saccades. Here, we develop a probabilistic model in the Generalized Linear Model framework to characterize and predict the...
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.