Search results for "NEURAL NETWORK"
showing 10 items of 1385 documents
Reconstruction of Micropattern Detector Signals using Convolutional Neural Networks
2017
Micropattern gaseous detector (MPGD) technologies, such as GEMs or MicroMegas, are particularly suitable for precision tracking and triggering in high rate environments. Given their relatively low production costs, MPGDs are an exemplary candidate for the next generation of particle detectors. Having acknowledged these advantages, both the ATLAS and CMS collaborations at the LHC are exploiting these new technologies for their detector upgrade programs in the coming years. When MPGDs are utilized for triggering purposes, the measured signals need to be precisely reconstructed within less than 200 ns, which can be achieved by the usage of FPGAs. In this work, we present a novel approach to id…
Effect of the milling conditions on the degree of amorphization of selenium by milling in a planetary ball mill
2007
The effect of the milling parameters (rotation speed of the milling device and duration of milling) on the phase composition of the products of milling of fully crystalline selenium has been investigated. The milling was conducted using a planetary micromill and the phase composition of the milling products was determined by differential thermal analysis. It has been found that ball milling leads to the partial amorphization of the starting crystalline material. The content of amorphous phase in the milling products depends, in a rather complicated way, on the milling parameters. At the milling parameters adopted in the present study, the milling product was never fully amorphous. The compl…
Devitrification of the Kob-Andersen glass former: Competition with the locally favored structure
2018
Abstract Supercooled liquids are kinetically trapped materials in which the transition to a thermodynamically more stable state with long-range order is strongly suppressed. To assess the glass-forming abilities of a liquid empirical rules exist, but a comprehensive microscopic picture of devitrification is still missing. Here we study the crystallization of a popular model glass former, the binary Kob-Andersen mixture, in small systems. We perform trajectory sampling employing the population of the locally favored structure as order parameter. While for large population a dynamical phase transition has been reported, here we show that biasing towards a small population of locally favored s…
Mapping discounted and undiscounted Markov Decision Problems onto Hopfield neural networks
1995
This paper presents a framework for mapping the value-iteration and related successive approximation methods for Markov Decision Problems onto Hopfield neural networks, for both discounted and undiscounted versions of the finite state and action spaces. We analyse the asymptotic behaviour of the control sets and we give some estimates on the convergence rate for the value-iteration scheme. We relate the convergence properties on an energy function which represents the key point in mapping Markov Decision Problems onto Hopfield networks. Finally, an application from queueing systems in communication networks is taken into consideration and the results of computer simulation of Hopfield netwo…
Critical phenomena without “hyper scaling”: How is the finite-size scaling analysis of Monte Carlo data affected?
2010
Abstract The finite size scaling analysis of Monte Carlo data is discussed for two models for which hyperscaling is violated: (i) the random field Ising model (using a model for a colloid-polymer mixture in a random matrix as a representative) (ii) The Ising bi-pyramid in computing surface fields.
A Hybrid Neural Network Architecture for Dynamic Scene Understanding
1997
A hyprdid (neural and symbolic) architecture allowing for a deep understanding of moving scenes is described. The architecture is based on a working and effective integration among three levels of representation of data coming out from external sensors.
Strategies investigation in using artificial neural network for landslide susceptibility mapping: application to a Sicilian catchment
2013
Susceptibility assessment of areas prone to landsliding remains one of the most useful approaches in landslide hazard analysis. The key point of such analysis is the correlation between the physical phenomenon and its triggering factors based on past observations. Many methods have been developed in the scientific literature to capture and model this correlation, usually within a geographic information system (GIS) framework. Among these, the use of neural networks, in particular the multi-layer perceptron (MLP) networks, has provided successful results. A successful application of the MLP method to a basin area requires the definition of different model strategies, such as the sample selec…
Towards Model-Based Reinforcement Learning for Industry-Near Environments
2019
Deep reinforcement learning has over the past few years shown great potential in learning near-optimal control in complex simulated environments with little visible information. Rainbow (Q-Learning) and PPO (Policy Optimisation) have shown outstanding performance in a variety of tasks, including Atari 2600, MuJoCo, and Roboschool test suite. Although these algorithms are fundamentally different, both suffer from high variance, low sample efficiency, and hyperparameter sensitivity that, in practice, make these algorithms a no-go for critical operations in the industry.
Description of Hysteresis in Lithium Battery by Classical Preisach Model
2012
In this paper Preisach Model is applied to obtain a mathematical model of the hysteresis in lithium battery. Preisach Model allows to describe the hysteresis of charging and discharging cycles in a lithium battery. The identification of the model is obtained by using a neural network technique developed for magnetic systems. The model is verified on some experimental tests on commercial batteries.
Identification of parameters of the Jiles-Atherton model by neural networks
2011
In this paper a procedure for the identification of the parameters of the Jiles–Atherton (JA) model is presented. The parameters of the JA model of a material are found by using a neural network trained by a collection of hysteresis curves, whose parameters are known. After a presentation of the Jiles–Atherton model, the neural network and the training procedure are described and the method is validated by using some numerical, as well as experimental, data.