scholarly journals Optimising simulations for diphoton production at hadron colliders using amplitude neural networks

2021 ◽  
Vol 2021 (8) ◽  
Author(s):  
Joseph Aylett-Bullock ◽  
Simon Badger ◽  
Ryan Moodie
Keyword(s):  
Neural Networks ◽  
Hadron Collider ◽  
Gluon Fusion ◽  
Simulation Method ◽  
Monte Carlo Event ◽  
Event Generator ◽  
Learning Technology ◽  
Scattering Problems ◽  
Network Simulations ◽  
The One

Abstract Machine learning technology has the potential to dramatically optimise event generation and simulations. We continue to investigate the use of neural networks to approximate matrix elements for high-multiplicity scattering processes. We focus on the case of loop-induced diphoton production through gluon fusion, and develop a realistic simulation method that can be applied to hadron collider observables. Neural networks are trained using the one-loop amplitudes implemented in the NJet C++ library, and interfaced to the Sherpa Monte Carlo event generator, where we perform a detailed study for 2 → 3 and 2 → 4 scattering problems. We also consider how the trained networks perform when varying the kinematic cuts effecting the phase space and the reliability of the neural network simulations.

scholarly journals Automatic Detection of Arrhythmia Based on Multi-Resolution Representation of ECG Signal

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1579
Author(s):  
Dongqi Wang ◽  
Qinghua Meng ◽  
Dongming Chen ◽  
Hupo Zhang ◽  
Lisheng Xu
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Channel Model ◽  
Expert Knowledge ◽  
Automatic Detection ◽  
Data Representation ◽  
Learning Technology ◽  
Arrhythmia Detection ◽  
Automatic Feature Extraction

Automatic detection of arrhythmia is of great significance for early prevention and diagnosis of cardiovascular disease. Traditional feature engineering methods based on expert knowledge lack multidimensional and multi-view information abstraction and data representation ability, so the traditional research on pattern recognition of arrhythmia detection cannot achieve satisfactory results. Recently, with the increase of deep learning technology, automatic feature extraction of ECG data based on deep neural networks has been widely discussed. In order to utilize the complementary strength between different schemes, in this paper, we propose an arrhythmia detection method based on the multi-resolution representation (MRR) of ECG signals. This method utilizes four different up to date deep neural networks as four channel models for ECG vector representations learning. The deep learning based representations, together with hand-crafted features of ECG, forms the MRR, which is the input of the downstream classification strategy. The experimental results of big ECG dataset multi-label classification confirm that the F1 score of the proposed method is 0.9238, which is 1.31%, 0.62%, 1.18% and 0.6% higher than that of each channel model. From the perspective of architecture, this proposed method is highly scalable and can be employed as an example for arrhythmia recognition.

scholarly journals Deep Neural Networks Classification via Binary Error-Detecting Output Codes

2021 ◽  
Vol 11 (8) ◽  
pp. 3563
Author(s):  
Martin Klimo ◽  
Peter Lukáč ◽  
Peter Tarábek
Keyword(s):  
Neural Networks ◽  
Categorical Data ◽  
Coding Theory ◽  
Hamming Distance ◽  
Activation Function ◽  
Block Codes ◽  
Ease Of Use ◽  
Linear Block Codes ◽  
Binary Coding ◽  
The One

One-hot encoding is the prevalent method used in neural networks to represent multi-class categorical data. Its success stems from its ease of use and interpretability as a probability distribution when accompanied by a softmax activation function. However, one-hot encoding leads to very high dimensional vector representations when the categorical data’s cardinality is high. The Hamming distance in one-hot encoding is equal to two from the coding theory perspective, which does not allow detection or error-correcting capabilities. Binary coding provides more possibilities for encoding categorical data into the output codes, which mitigates the limitations of the one-hot encoding mentioned above. We propose a novel method based on Zadeh fuzzy logic to train binary output codes holistically. We study linear block codes for their possibility of separating class information from the checksum part of the codeword, showing their ability not only to detect recognition errors by calculating non-zero syndrome, but also to evaluate the truth-value of the decision. Experimental results show that the proposed approach achieves similar results as one-hot encoding with a softmax function in terms of accuracy, reliability, and out-of-distribution performance. It suggests a good foundation for future applications, mainly classification tasks with a high number of classes.

FAST SPRINGBACK SIMULATION OF BENDING FORMING BASED ON ONE-STEP INVERSE ANALYSIS

2008 ◽  
Vol 22 (31n32) ◽  
pp. 6057-6063
Author(s):  
YI-DONG BAO ◽  
WEN-LIANG CHEN ◽  
HONG WU
Keyword(s):  
Sheet Metal Forming ◽  
Metal Forming ◽  
Inverse Analysis ◽  
Simulation Method ◽  
Spring Back ◽  
Deformation Path ◽  
Bending Process ◽  
One Step ◽  
Implicit Solver ◽  
The One

A simplified one-step inverse analysis of sheet metal forming is a suitable tool to simulate the bending forming since the deformation path of bending forming is an approximately proportion one. A fast spring-back simulation method based on one-step analysis is proposed. First, the one-step inverse analysis is applied to obtain the stress distribution at the final stage of bending. Then, the unloading to get a spring back is simulated by LS-DYNA implicit solver. These processes are applied to the unconstrained cylindrical bending and the truck member rail. The spring-back and member rail widths at the several key sections are compared with experimental ones. It is well demonstrated that the proposed method is an effective way to predict the spring-back by unloading after bending process.

kronos: A Monte Carlo event generator for higher order QED corrections at HERA — Status report

1992 ◽  
Vol 29 (1) ◽  
pp. 247-252
Author(s):  
Harald Anlauf ◽  
Panagiotis Manakos ◽  
Thomas Mannel ◽  
Hans D. Dahmen ◽  
Thorsten Ohl
Keyword(s):  
Monte Carlo ◽  
Higher Order ◽  
Monte Carlo Event ◽  
Status Report ◽  
Event Generator ◽  
Monte Carlo Event Generator

Modeling of Distributed Parameter Processes with Neural Networks

2014 ◽  
Vol 555 ◽  
pp. 530-540
Author(s):  
Honoriu Vălean ◽  
Mihail Abrudean ◽  
Mihaela Ligia Ungureşan ◽  
Iulia Clitan ◽  
Vlad Mureşan
Keyword(s):  
Neural Networks ◽  
Simulation Method ◽  
Distributed Parameter ◽  
Hearth Furnace ◽  
Original Solution ◽  
Steel Pipes ◽  
The Matrix ◽  
The Neural Networks ◽  
Derivatives Of ◽  
Seamless Steel

In this paper an original solution for the modeling of distributed parameter processes using neural networks is presented. The proposed method represents a particular alternative to a very accurate modeling-simulation method for this kind of processes, the method based on the matrix of partial derivatives of the state vector (Mpdx), associated with Taylor series. In order to compare the performances generated by the two methods, a distributed parameter thermal process associated to a rotary hearth furnace (R.H.F) from the technological flow of producing seamless steel pipes is considered. The main similarities and differences between the two methods are highlighted in the paper. The treated solution represents a premise for the usage of the neural networks in the automatic control of the distributed parameter processes domain.

scholarly journals Investigation of the influence of nuclear matter on hard neutrino nucleus interaction using the HARDPING Monte Carlo Event Generator

2016 ◽  
Vol 125 ◽  
pp. 04021
Author(s):  
Ya.A. Berdnikov ◽  
A.Ya. Berdnikov ◽  
V.T. Kim ◽  
A.E. Ivanov ◽  
D.P. Suetin ◽  
...  
Keyword(s):  
Monte Carlo ◽  
Nuclear Matter ◽  
Monte Carlo Event ◽  
Event Generator ◽  
Nucleus Interaction ◽  
Monte Carlo Event Generator

Autoencoder for ecg signal outlier processing in system of biometric authentication

2019 ◽  
Vol 24 (1-2) ◽  
pp. 108-117
Author(s):  
Khoma V.V. ◽  
◽  
Khoma Y.V. ◽  
Khoma P.P. ◽  
Sabodashko D.V. ◽  
...  
Keyword(s):  
Neural Networks ◽  
Statistical Approach ◽  
Main Idea ◽  
Biometric Identification ◽  
Ecg Signal ◽  
Biometric Authentication ◽  
Lower Accuracy ◽  
Novel Method ◽  
Waveform Distortions ◽  
The One

A novel method for ECG signal outlier processing based on autoencoder neural networks is presented in the article. Typically, heartbeats with serious waveform distortions are treated as outliers and are skipped from the authentication pipeline. The main idea of the paper is to correct these waveform distortions rather them in order to provide the system with better statistical base. During the experiments, the optimum autoencoder architecture was selected. An open Physionet ECGID database was used to verify the proposed method. The results of the studies were compared with previous studies that considered the correction of anomalies based on a statistical approach. On the one hand, the autoencoder shows slightly lower accuracy than the statistical method, but it greatly simplifies the construction of biometric identification systems, since it does not require precise tuning of hyperparameters.

Revealing the importance of international and domestic cooperation by using artificial neural networks: case of European radical and incremental innovators

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Viktor Prokop ◽  
Jan Stejskal ◽  
Beata Mikusova Merickova ◽  
Samuel Amponsah Odei
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
International Cooperation ◽  
Incremental Innovation ◽  
Content Type ◽  
Innovation Activities ◽  
Design Activities ◽  
Artificial Neural ◽  
The One ◽  
Innovation Processes

PurposeThe purpose of this study is to introduce innovative ideas into the treatment of the radical and incremental innovations and to fill the research gap by using: (1) methods that can perform complicated tasks and solve complex problems leading in creation of radical and incremental innovation and (2) a broad sample of firms across countries. The authors’ ambition is to contribute to the scientific knowledge by producing evidence about the novel usage of artificial neural network techniques for measuring European firms' innovation activities appearing in black boxes of innovation processes.Design/methodology/approachIn this study, the authors incorporate an international context into Chesbrough's open innovation (OI) theory and, on the one hand, support the hypothesis that European radical innovators benefit more from foreign cooperation than incremental innovators. On the other hand, the results of the analyses show that European incremental innovators rely on domestic cooperation supported by cooperation with foreign public research institutes. Moreover, the use of decision trees (DT) allows the authors to reveal specific patterns of successful innovators emerging within the hidden layers of neural networks.FindingsThe authors prove that radical European innovators using either internal or external R&D strategies, while the combinations of these strategies do not bring successful innovation outputs. In contrast, European incremental innovators benefit from various internal R&D processes in which engagement in design activities plays a crucial role.Originality/valueThe authors introduce innovative ideas into the treatment of hidden innovation processes and measuring the innovation performance (affected by domestic or international cooperation) of European firms. The approach places emphasis on the novelty of innovation and the issue of international cooperation in the era of OI by designing the framework using a combination of artificial neural networks and DT.

MODELING OF NEURAL PREDICTIVE CONTROL OF IRRIGATION MACHINES

2021 ◽  
pp. 14-22
Author(s):  
G. N. KAMYSHOVA ◽  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Real Time ◽  
Predictive Control ◽  
Intelligent Control ◽  
Machine Learning Algorithms ◽  
Intelligent Control Systems ◽  
Artificial Neural ◽  
The One ◽  
Neural Predictive Control

The purpose of the study is to develop new scientific approaches to improve the efficiency of irrigation machines. Modern digital technologies allow the collection of data, their analysis and operational management of equipment and technological processes, often in real time. All this allows, on the one hand, applying new approaches to modeling technical systems and processes (the so-called “data-driven models”), on the other hand, it requires the development of fundamentally new models, which will be based on the methods of artificial intelligence (artificial neural networks, fuzzy logic, machine learning algorithms and etc.).The analysis of the tracks and the actual speeds of the irrigation machines in real time showed their significant deviations in the range from the specified speed, which leads to a deterioration in the irrigation parameters. We have developed an irrigation machine’s control model based on predictive control approaches and the theory of artificial neural networks. Application of the model makes it possible to implement control algorithms with predicting the response of the irrigation machine to the control signal. A diagram of an algorithm for constructing predictive control, a structure of a neuroregulator and tools for its synthesis using modern software are proposed. The versatility of the model makes it possible to use it both to improve the efficiency of management of existing irrigation machines and to develop new ones with integrated intelligent control systems.

Sign in / Sign up

Export Citation Format

Share Document