Search results for "NEURAL NETWORK"

Tricritical universality in a two-dimensional spin fluid

1995

Monte Carlo simulations are used to investigate the tricritical point properties of a 2d spin fluid. Measurements of the scaling operator distributions are employed in conjunction with a finite-size scaling analysis to locate the tricritical point and determine the directions of the relevant scaling fields and their associated tricritical exponents. The scaling operator distributions and exponents are shown to match quantitatively those of the 2d Blume-Capel model, confirming that both models belong to the same universality class. Mean-field calculations of the tricritical point properties are also compared with the simulation measurements.

A discrimination technique for extensive air showers based on multiscale, lacunarity and neural network analysis

2011

We present a new method for the identification of extensive air showers initiated by different primaries. The method uses the multiscale concept and is based on the analysis of multifractal behaviour and lacunarity of secondary particle distributions together with a properly designed and trained artificial neural network. In the present work the method is discussed and applied to a set of fully simulated vertical showers, in the experimental framework of ARGO-YBJ, to obtain hadron to gamma primary separation. We show that the presented approach gives very good results, leading, in the 1–10 TeV energy range, to a clear improvement of the discrimination power with respect to the existing figu…

Numerical simulations and exactly soluble spin-glass models.

1985

Some general arguments based on recent numerical work are presented to explain the different behavior of short-range, random-bond and long-range, random-site spin glasses. We then analyze an exactly soluble spin-glass model, which may be solved without replicas, and show that, except for the absence of microscopic metastable states, its main features are consistent with the long-range picture.

CNN-based People Detection in Voxel Space using Intensity Measurements and Point Cluster Flattening

2021

In this paper real-time people detection is demonstrated in a relatively large indoor industrial robot cell as well as in an outdoor environment. Six depth sensors mounted at the ceiling are used to generate a merged point cloud of the cell. The merged point cloud is segmented into clusters and flattened into gray-scale 2D images in the xy and xz planes. These images are then used as input to a classifier based on convolutional neural networks (CNNs). The final output is the 3D position (x,y,z) and bounding box representing the human. The system is able to detect and track multiple humans in real-time, both indoors and outdoors. The positional accuracy of the proposed method has been verifi…

Transient behavior of a population dynamical model

2005

The transient behavior of an ecosystem with N random interacting species in the presence of a multiplicative noise is analyzed. The multiplicative noise mimics the interaction with the environment. We investigate different asymptotic dynamical regimes and the role of the external noise on the probability distribution of the local field.

Corrigendum: Spectral Entropy Based Neuronal Network Synchronization Analysis Based on Microelectrode Array Measurements

2020

A neural network clustering algorithm for the ATLAS silicon pixel detector

2014

A novel technique to identify and split clusters created by multiple charged particles in the ATLAS pixel detector using a set of artificial neural networks is presented. Such merged clusters are a common feature of tracks originating from highly energetic objects, such as jets. Neural networks are trained using Monte Carlo samples produced with a detailed detector simulation. This technique replaces the former clustering approach based on a connected component analysis and charge interpolation. The performance of the neural network splitting technique is quantified using data from proton-proton collisions at the LHC collected by the ATLAS detector in 2011 and from Monte Carlo simulations. …

A novel arousal-based individual screening reveals susceptibility and resilience to PTSD-like phenotypes in mice

2021

Translational animal models for studying post-traumatic stress disorder (PTSD) are valuable for elucidating the poorly understood neurobiology of this neuropsychiatric disorder. These models should encompass crucial features, including persistence of PTSD-like phenotypes triggered after exposure to a single traumatic event, trauma susceptibility/resilience and predictive validity. Here we propose a novel arousal-based individual screening (AIS) model that recapitulates all these features. The AIS model was designed by coupling the traumatization (24 h restraint) of C57BL/6 J mice with a novel individual screening. This screening consists of z-normalization of post-trauma changes in startle …

Machine Learning Identification of Pro-arrhythmic Structures in Cardiac Fibrosis

2021

Cardiac fibrosis and other scarring of the heart, arising from conditions ranging from myocardial infarction to ageing, promotes dangerous arrhythmias by blocking the healthy propagation of cardiac excitation. Owing to the complexity of the dynamics of electrical signalling in the heart, however, the connection between different arrangements of blockage and various arrhythmic consequences remains poorly understood. Where a mechanism defies traditional understanding, machine learning can be invaluable for enabling accurate prediction of quantities of interest (measures of arrhythmic risk) in terms of predictor variables (such as the arrangement or pattern of obstructive scarring). In this st…

Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks

2016

This paper proposes Markovian Generative Adversarial Networks (MGANs), a method for training generative networks for efficient texture synthesis. While deep neural network approaches have recently demonstrated remarkable results in terms of synthesis quality, they still come at considerable computational costs (minutes of run-time for low-res images). Our paper addresses this efficiency issue. Instead of a numerical deconvolution in previous work, we precompute a feed-forward, strided convolutional network that captures the feature statistics of Markovian patches and is able to directly generate outputs of arbitrary dimensions. Such network can directly decode brown noise to realistic textu…