Identification of new long-lived particles using deep neural networks

We present the development of a deep neural network for identifying generic displaced jets arising from the decays of exotic long-lived particles in data recorded by the CMS detector at the CERN LHC. Various jet features including detailed information about each clustered particle candidate as well as reconstructed secondary vertices are refined through the use of 1-dimensional convolution layers before being combined with high-level engineered features and passed through a series of fully-connected layers. The proper lifetime of the long-lived particle, cτ0, is treated as a parameter of the neural network model, which allows for hypothesis testing over several orders of magnitude ranging from cτ0 = 1 µm to 10 m. Domain adaptation by backward propagation is performed to construct domain-independent features at an intermediate layer of the network to mitiage difference between simulation and data. The training is performed by streaming ROOT trees containing O(100M) jets directly into the TensorFlow queue system, which allows for a flexible selection of input features and asynchronous preprocessing. The application of the tagger is showcased in a search for long-lived gluinos as predicted by split supersymmetric models demonstrating significant gains in sensitivity over a reference analysis.

Download Full-text

Adaptive Control Structure with Neural Data Processing Applied for Electrical Drive with Elastic Shaft

Energies ◽

10.3390/en14123389 ◽

2021 ◽

Vol 14 (12) ◽

pp. 3389

Author(s):

Marcin Kamiński ◽

Krzysztof Szabat

Keyword(s):

Neural Network ◽

Adaptive Control ◽

Dynamic Properties ◽

Experimental Studies ◽

Drive System ◽

Particle Swarm Optimizer ◽

The Neural Network ◽

Control Concept ◽

Constant Coefficients ◽

Selection Of

This paper presents issues related to the adaptive control of the drive system with an elastic clutch connecting the main motor and the load machine. Firstly, the problems and the main algorithms often implemented for the mentioned object are analyzed. Then, the control concept based on the RNN (recurrent neural network) for the drive system with the flexible coupling is thoroughly described. For this purpose, an adaptive model inspired by the Elman model is selected, which is related to internal feedback in the neural network. The indicated feature improves the processing of dynamic signals. During the design process, for the selection of constant coefficients of the controller, the PSO (particle swarm optimizer) is applied. Moreover, in order to obtain better dynamic properties and improve work in real conditions, one model based on the ADALINE (adaptive linear neuron) is introduced into the structure. Details of the algorithm used for the weights’ adaptation are presented (including stability analysis) to perform the shaft torque signal filtering. The effectiveness of the proposed approach is examined through simulation and experimental studies.

Download Full-text

Real-Time Adversarial Attack Detection with Deep Image Prior Initialized as a High-Level Representation Based Blurring Network

Electronics ◽

10.3390/electronics10010052 ◽

2020 ◽

Vol 10 (1) ◽

pp. 52

Author(s):

Richard Evan Sutanto ◽

Sukho Lee

Keyword(s):

Neural Network ◽

Attack Detection ◽

Detection Methods ◽

Defense System ◽

Image Prior ◽

The Neural Network ◽

Adversarial Examples ◽

Deep Image ◽

Adversarial Attack ◽

High Level

Several recent studies have shown that artificial intelligence (AI) systems can malfunction due to intentionally manipulated data coming through normal channels. Such kinds of manipulated data are called adversarial examples. Adversarial examples can pose a major threat to an AI-led society when an attacker uses them as means to attack an AI system, which is called an adversarial attack. Therefore, major IT companies such as Google are now studying ways to build AI systems which are robust against adversarial attacks by developing effective defense methods. However, one of the reasons why it is difficult to establish an effective defense system is due to the fact that it is difficult to know in advance what kind of adversarial attack method the opponent is using. Therefore, in this paper, we propose a method to detect the adversarial noise without knowledge of the kind of adversarial noise used by the attacker. For this end, we propose a blurring network that is trained only with normal images and also use it as an initial condition of the Deep Image Prior (DIP) network. This is in contrast to other neural network based detection methods, which require the use of many adversarial noisy images for the training of the neural network. Experimental results indicate the validity of the proposed method.

Download Full-text

Machine learning estimation of tissue optical properties

Scientific Reports ◽

10.1038/s41598-021-85994-w ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Brett H. Hokr ◽

Joel N. Bixler

Keyword(s):

Neural Network ◽

Optical Properties ◽

Biological Tissues ◽

Homogeneous Layer ◽

In Vivo Measurement ◽

Tissue Optical Properties ◽

The Neural Network ◽

Spatio Temporal ◽

Fully Connected

AbstractDynamic, in vivo measurement of the optical properties of biological tissues is still an elusive and critically important problem. Here we develop a technique for inverting a Monte Carlo simulation to extract tissue optical properties from the statistical moments of the spatio-temporal response of the tissue by training a 5-layer fully connected neural network. We demonstrate the accuracy of the method across a very wide parameter space on a single homogeneous layer tissue model and demonstrate that the method is insensitive to parameter selection of the neural network model itself. Finally, we propose an experimental setup capable of measuring the required information in real time in an in vivo environment and demonstrate proof-of-concept level experimental results.

Download Full-text

Estimation of Site Amplification from Geotechnical Array Data Using Neural Networks

Bulletin of the Seismological Society of America ◽

10.1785/0120200346 ◽

2021 ◽

Author(s):

Daniel Roten ◽

Kim B. Olsen

Keyword(s):

Neural Network ◽

Neural Networks ◽

Site Response ◽

Site Amplification ◽

Data Driven ◽

Vertical Array ◽

The Neural Network ◽

Simplifying Assumptions ◽

The Mean ◽

Fully Connected

ABSTRACT We use deep learning to predict surface-to-borehole Fourier amplification functions (AFs) from discretized shear-wave velocity profiles. Specifically, we train a fully connected neural network and a convolutional neural network using mean AFs observed at ∼600 KiK-net vertical array sites. Compared with predictions based on theoretical SH 1D amplifications, the neural network (NN) results in up to 50% reduction of the mean squared log error between predictions and observations at sites not used for training. In the future, NNs may lead to a purely data-driven prediction of site response that is independent of proxies or simplifying assumptions.

Download Full-text

Prediction of Mining Conditions in Geotechnically Complex Sites

Mining ◽

10.3390/mining1030018 ◽

2021 ◽

Vol 1 (3) ◽

pp. 279-296

Author(s):

Marc Elmouttie ◽

Jane Hodgkinson ◽

Peter Dean

Keyword(s):

Neural Network ◽

Historical Data ◽

Synthetic Data ◽

Mining Operations ◽

Body Of Knowledge ◽

Future Production ◽

The Neural Network ◽

Mining Conditions ◽

Parameter Interactions ◽

High Level

Geotechnical complexity in mining often leads to geotechnical uncertainty which impacts both safety and productivity. However, as mining progresses, particularly for strip mining operations, a body of knowledge is acquired which reduces this uncertainty and can potentially be used by mining engineers to improve the prediction of future mining conditions. In this paper, we describe a new method to support this approach based on modelling and neural networks. A high-level causal model of the mining operations based on historical data for a number of parameters was constructed which accounted for parameter interactions, including hydrogeological conditions, weather, and prior operations. An artificial neural network was then trained on this historical data, including production data. The network can then be used to predict future production based on presently observed mining conditions as mining proceeds and compared with the model predictions. Agreement with the predictions indicates confidence that the neural network predictions are properly supported by the newly available data. The efficacy of this approach is demonstrated using semi-synthetic data based on an actual mine.

Download Full-text

APPLICATION OF NEURAL NETWORK TECHNOLOGIES IN THE DENTAL CARIES FORECAST

Wiadomości Lekarskie ◽

10.36740/wlek202007135 ◽

2020 ◽

Vol 73 (7) ◽

pp. 1499-1504

Author(s):

Oleksandr A. Udod ◽

Hanna S. Voronina ◽

Olena Yu. Ivchenkova

Keyword(s):

Neural Network ◽

Dental Caries ◽

Old Patients ◽

Software Application ◽

The Neural Network ◽

Software Product ◽

High Level ◽

Network Technologies ◽

Python Programming ◽

And Training

The aim: of the work was to develop and apply in the clinical trial a software product for the dental caries prediction based on neural network programming. Materials and methods: Dental examination of 73 persons aged 6-7, 12-15 and 35-44 years was carried out. The data obtained during the survey were used as input for the construction and training of the neural network. The output index was determined by the increase in the intensity of caries, taking into account the number of cavities. To build a neural network, a high-level Python programming language with the NumPay extension was used. Results: The intensity of carious dental lesions was the highest in 35-44 years old patients – 6.69 ± 0.38, in 6-7 years old children and 12-15 years old children it was 3.85 ± 0.27 and 2.15 ± 0.24, respectively (p <0.05). After constructing and training the neural network, 61 true and 12 false predictions were obtained based on these indices, the accuracy of predicting the occurrence of caries was 83.56%. Based on these results, a graphical user interface for the “CariesPro” software application was created. Conclusions: The resulting neural network and the software product based on it permit to predict the development of dental caries in persons of all ages with a probability of 83.56%.

Download Full-text

NEUROSLAB – neural network system for horizontal formwork selection

Canadian Journal of Civil Engineering ◽

10.1139/l95-088 ◽

1995 ◽

Vol 22 (4) ◽

pp. 785-792 ◽

Cited By ~ 3

Author(s):

Awad S. Hanna ◽

Ahmed B. Senouci

Keyword(s):

Neural Network ◽

Expert System ◽

Back Propagation ◽

Network System ◽

Neural Network System ◽

Testing Error ◽

Training Error ◽

Training Examples ◽

High Level ◽

Selection Of

This paper presents an overview of the neural network technique as a tool for concrete formwork selection. The paper discusses the development and the implementation of a neural network system, NEUROSLAB, for the selection of horizontal formwork systems. A rule-based expert system for the selection of horizontal systems, SLABFORM, was used as the basis for the development of NEUROSLAB. A training set of 202 cases was used to train the network. The network adequately learned the training examples with an average training error of 0.025. A set of 50 cases was used to test the generalization ability of the system. The network was able to accurately select the appropriate horizontal formwork system with an average testing error of 0.057. The ability of the network to deal with noisy data was also tested. Up to 50% noise was added to the data and introduced to the network. The results showed that the network presented could accurately identify the appropriate horizontal formwork system at high level of noise. Finally, the solution chosen by an expert was compared to that produced by the network. The network was able to mimic the expert's formwork selection. Key words: formwork, horizontal formwork systems, neural network, formwork selection, back propagation, expert system.

Download Full-text

Methods for the selection of parameters and structure of the neural network model

Semi-Empirical Neural Network Modeling and Digital Twins Development ◽

10.1016/b978-0-12-815651-3.00003-1 ◽

2020 ◽

pp. 73-103

Author(s):

Dmitriy Tarkhov ◽

Alexander Vasilyev

Keyword(s):

Neural Network ◽

Network Model ◽

Neural Network Model ◽

The Neural Network ◽

Selection Of

Download Full-text

Fitting potential energy surfaces to sum-of-products form with neural networks using exponential neurons

Journal of Theoretical and Computational Chemistry ◽

10.1142/s0219633617300014 ◽

2017 ◽

Vol 16 (05) ◽

pp. 1730001 ◽

Cited By ~ 6

Author(s):

Alex Brown ◽

E. Pradhan

Keyword(s):

Neural Network ◽

Potential Energy ◽

Ab Initio ◽

Potential Energy Surfaces ◽

Quantum Dynamics ◽

The Neural Network ◽

Sum Of Products ◽

High Level ◽

Energy Surfaces ◽

Dynamics Problems

In this paper, the use of the neural network (NN) method with exponential neurons for directly fitting ab initio data to generate potential energy surfaces (PESs) in sum-of-product form will be discussed. The utility of the approach will be highlighted using fits of CS2, HFCO, and HONO ground state PESs based upon high-level ab initio data. Using a generic interface between the neural network PES fitting, which is performed in MATLAB, and the Heidelberg multi-configuration time-dependent Hartree (MCTDH) software package, the PESs have been tested via comparison of vibrational energies to experimental measurements. The review demonstrates the potential of the PES fitting method, combined with MCTDH, to tackle high-dimensional quantum dynamics problems.

Download Full-text