scholarly journals Neural Network Decoders for Large-Distance 2D Toric Codes

Quantum ◽  
2020 ◽  
Vol 4 ◽  
pp. 310 ◽  
Author(s):  
Xiaotong Ni
Keyword(s):  
Neural Network ◽  
Minimum Weight ◽  
Code Distance ◽  
Matching Algorithm ◽  
Training Time ◽  
The Neural Network ◽  
Wide Range ◽  
Final Design ◽  
Bit Flip ◽  
Noise Models

We still do not have perfect decoders for topological codes that can satisfy all needs of different experimental setups. Recently, a few neural network based decoders have been studied, with the motivation that they can adapt to a wide range of noise models, and can easily run on dedicated chips without a full-fledged computer. The later feature might lead to fast speed and the ability to operate at low temperatures. However, a question which has not been addressed in previous works is whether neural network decoders can handle 2D topological codes with large distances. In this work, we provide a positive answer for the toric code \cite{Kitaev2003Faulttolerantanyon}. The structure of our neural network decoder is inspired by the renormalization group decoder \cite{duclos2010fast, duclos2013fault}. With a fairly strict policy on training time, when the bit-flip error rate is lower than 9% and syndrome extraction is perfect, the neural network decoder performs better when code distance increases. With a less strict policy, we find it is not hard for the neural decoder to achieve a performance close to the minimum-weight perfect matching algorithm. The numerical simulation is done up to code distance d=64. Last but not least, we describe and analyze a few failed approaches. They guide us to the final design of our neural decoder, but also serve as a caution when we gauge the versatility of stock deep neural networks. The source code of our neural decoder can be found at \cite{sourcecodegithub}.

Robust sideslip angle observer of commercial vehicles based on cornering stiffness estimation using neural network

2022 ◽  
pp. 095440702110724
Author(s):  
Chenyu Zhou ◽  
Liangyao Yu ◽  
Yong Li ◽  
Jian Song
Keyword(s):  
Neural Network ◽  
Mathematical Model ◽  
Stability Control ◽  
Estimation Accuracy ◽  
Accurate Estimation ◽  
Commercial Vehicles ◽  
Sideslip Angle ◽  
The Neural Network ◽  
Wide Range ◽  
The Mathematical Model

Accurate estimation of sideslip angle is essential for vehicle stability control. For commercial vehicles, the estimation of sideslip angle is challenging due to severe load transfer and tire nonlinearity. This paper presents a robust sideslip angle observer of commercial vehicles based on identification of tire cornering stiffness. Since tire cornering stiffness of commercial vehicles is greatly affected by tire force and road adhesion coefficient, it cannot be treated as a constant. To estimate the cornering stiffness in real time, the neural network model constructed by Levenberg-Marquardt backpropagation (LMBP) algorithm is employed. LMBP is a fast convergent supervised learning algorithm, which combines the steepest descent method and gauss-newton method, and is widely used in system parameter estimation. LMBP does not rely on the mathematical model of the actual system when building the neural network. Therefore, when the mathematical model is difficult to establish, LMBP can play a very good role. Considering the complexity of tire modeling, this study adopted LMBP algorithm to estimate tire cornering stiffness, which have simplified the tire model and improved the estimation accuracy. Combined with neural network, A time-varying Kalman filter (TVKF) is designed to observe the sideslip angle of commercial vehicles. To validate the feasibility of the proposed estimation algorithm, multiple driving maneuvers under different road surface friction have been carried out. The test results show that the proposed method has better accuracy than the existing algorithm, and it’s robust over a wide range of driving conditions.

Signal processing algorithm for neural networks with integrodifferential splines as an activation function and its particular case of image classification

2021 ◽  
Author(s):  
T.K. Biryukova
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Image Classification ◽  
Activation Function ◽  
Experimental Comparison ◽  
Training Time ◽  
Operation Speed ◽  
The Neural Network ◽  
Linear Algebraic Equations ◽  
Network Operation

Classic neural networks suppose trainable parameters to include just weights of neurons. This paper proposes parabolic integrodifferential splines (ID-splines), developed by author, as a new kind of activation function (AF) for neural networks, where ID-splines coefficients are also trainable parameters. Parameters of ID-spline AF together with weights of neurons are vary during the training in order to minimize the loss function thus reducing the training time and increasing the operation speed of the neural network. The newly developed algorithm enables software implementation of the ID-spline AF as a tool for neural networks construction, training and operation. It is proposed to use the same ID-spline AF for neurons in the same layer, but different for different layers. In this case, the parameters of the ID-spline AF for a particular layer change during the training process independently of the activation functions (AFs) of other network layers. In order to comply with the continuity condition for the derivative of the parabolic ID-spline on the interval (x x0, n) , its parameters fi (i= 0,...,n) should be calculated using the tridiagonal system of linear algebraic equations: To solve the system it is necessary to use two more equations arising from the boundary conditions for specific problems. For exam- ple the values of the grid function (if they are known) in the points (x x0, n) may be used for solving the system above: f f x0 = ( 0) , f f xn = ( n) . The parameters Iii+1 (i= 0,...,n−1 ) are used as trainable parameters of neural networks. The grid boundaries and spacing of the nodes of ID-spline AF are best chosen experimentally. The optimal selection of grid nodes allows improving the quality of results produced by the neural network. The formula for a parabolic ID-spline is such that the complexity of the calculations does not depend on whether the grid of nodes is uniform or non-uniform. An experimental comparison of the results of image classification from the popular FashionMNIST dataset by convolutional neural 0, x< 0 networks with the ID-spline AFs and the well-known ReLUx( ) =AF was carried out. The results reveal that the usage x x, ≥ 0 of the ID-spline AFs provides better accuracy of neural network operation than the ReLU AF. The training time for two convolutional layers network with two ID-spline AFs is just about 2 times longer than with two instances of ReLU AF. Doubling of the training time due to complexity of the ID-spline formula is the acceptable price for significantly better accuracy of the network. Wherein the difference of an operation speed of the networks with ID-spline and ReLU AFs will be negligible. The use of trainable ID-spline AFs makes it possible to simplify the architecture of neural networks without losing their efficiency. The modification of the well-known neural networks (ResNet etc.) by replacing traditional AFs with ID-spline AFs is a promising approach to increase the neural network operation accuracy. In a majority of cases, such a substitution does not require to train the network from scratch because it allows to use pre-trained on large datasets neuron weights supplied by standard software libraries for neural network construction thus substantially shortening training time.

scholarly journals High-Resolution Sea Surface Temperature and Salinity in the Global Ocean from Raw Satellite Data

2019 ◽  
Vol 11 (19) ◽  
pp. 2191 ◽  
Author(s):  
Encarni Medina-Lopez ◽  
Leonardo Ureña-Fuentes
Keyword(s):  
Neural Network ◽  
High Resolution ◽  
Satellite Data ◽  
Google Earth ◽  
Sea Surface ◽  
Global Ocean ◽  
The Neural Network ◽  
Wide Range ◽  
Sentinel 2

The aim of this work is to obtain high-resolution values of sea surface salinity (SSS) and temperature (SST) in the global ocean by using raw satellite data (i.e., without any band data pre-processing or atmospheric correction). Sentinel-2 Level 1-C Top of Atmosphere (TOA) reflectance data is used to obtain accurate SSS and SST information. A deep neural network is built to link the band information with in situ data from different buoys, vessels, drifters, and other platforms around the world. The neural network used in this paper includes shortcuts, providing an improved performance compared with the equivalent feed-forward architecture. The in situ information used as input for the network has been obtained from the Copernicus Marine In situ Service. Sentinel-2 platform-centred band data has been processed using Google Earth Engine in areas of 100 m × 100 m. Accurate salinity values are estimated for the first time independently of temperature. Salinity results rely only on direct satellite observations, although it presented a clear dependency on temperature ranges. Results show the neural network has good interpolation and extrapolation capabilities. Test results present correlation coefficients of 82 % and 84 % for salinity and temperature, respectively. The most common error for both SST and SSS is 0.4 ∘ C and 0 . 4 PSU. The sensitivity analysis shows that outliers are present in areas where the number of observations is very low. The network is finally applied over a complete Sentinel-2 tile, presenting sensible patterns for river-sea interaction, as well as seasonal variations. The methodology presented here is relevant for detailed coastal and oceanographic applications, reducing the time for data pre-processing, and it is applicable to a wide range of satellites, as the information is directly obtained from TOA data.

scholarly journals Evaluation of the Dreicer runaway generation rate in the presence of high- impurities using a neural network

2019 ◽  
Vol 85 (6) ◽  
Author(s):  
L. Hesslow ◽  
L. Unnerfelt ◽  
O. Vallhagen ◽  
O. Embreus ◽  
M. Hoppe ◽  
...  
Keyword(s):  
Neural Network ◽  
Runaway Electron ◽  
The Self ◽  
Generation Rate ◽  
Integrated Modelling ◽  
Plasma Parameters ◽  
The Neural Network ◽  
Wide Range ◽  
Self Consistent ◽  
Computationally Expensive

Integrated modelling of electron runaway requires computationally expensive kinetic models that are self-consistently coupled to the evolution of the background plasma parameters. The computational expense can be reduced by using parameterized runaway generation rates rather than solving the full kinetic problem. However, currently available generation rates neglect several important effects; in particular, they are not valid in the presence of partially ionized impurities. In this work, we construct a multilayer neural network for the Dreicer runaway generation rate which is trained on data obtained from kinetic simulations performed for a wide range of plasma parameters and impurities. The neural network accurately reproduces the Dreicer runaway generation rate obtained by the kinetic solver. By implementing it in a fluid runaway-electron modelling tool, we show that the improved generation rates lead to significant differences in the self-consistent runaway dynamics as compared to the results using the previously available formulas for the runaway generation rate.

scholarly journals TRAINING NEURAL NETWORK FOR TAXI PASSENGER DEMAND FORECASTING USING GRAPHICS PROCESSING UNITS

2020 ◽  
Vol 2 (1) ◽  
pp. 29-36
Author(s):  
M. I. Zghoba ◽  
◽  
Yu. I. Hrytsiuk ◽  
Keyword(s):  
Neural Network ◽  
Graphics Processing Units ◽  
Neural Network Training ◽  
Training Time ◽  
Passenger Demand ◽  
Network Training ◽  
The Neural Network ◽  
Input Dataset ◽  
Speed Up ◽  
Graphics Processing

The peculiarities of neural network training for forecasting taxi passenger demand using graphics processing units are considered, which allowed to speed up the training procedure for different sets of input data, hardware configurations, and its power. It has been found that taxi services are becoming more accessible to a wide range of people. The most important task for any transportation company and taxi driver is to minimize the waiting time for new orders and to minimize the distance from drivers to passengers on order receiving. Understanding and assessing the geographical passenger demand that depends on many factors is crucial to achieve this goal. This paper describes an example of neural network training for predicting taxi passenger demand. It shows the importance of a large input dataset for the accuracy of the neural network. Since the training of a neural network is a lengthy process, parallel training was used to speed up the training. The neural network for forecasting taxi passenger demand was trained using different hardware configurations, such as one CPU, one GPU, and two GPUs. The training times of one epoch were compared along with these configurations. The impact of different hardware configurations on training time was analyzed in this work. The network was trained using a dataset containing 4.5 million trips within one city. The results of this study show that the training with GPU accelerators doesn't necessarily improve the training time. The training time depends on many factors, such as input dataset size, splitting of the entire dataset into smaller subsets, as well as hardware and power characteristics.

scholarly journals Development of a generative adversarial neural network for identification of potential HIV-1 inhibitors by deep learning methods

Informatics ◽  
2020 ◽  
Vol 17 (1) ◽  
pp. 7-17
Author(s):  
G. I. Nikolaev ◽  
N. A. Shuldov ◽  
A. I. Anishenko, ◽  
A. V. Tuzikov ◽  
A. M. Andrianov
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Rational Design ◽  
Binding Free Energy ◽  
Viral Envelope ◽  
Training Dataset ◽  
Learning Methods ◽  
The Neural Network ◽  
Wide Range ◽  
Hiv 1

A generative adversarial autoencoder for the rational design of potential HIV-1 entry inhibitors able to block the region of the viral envelope protein gp120 critical for the virus binding to cellular receptor CD4 was developed using deep learning methods. The research were carried out to create the  architecture of the neural network, to form  virtual compound library of potential anti-HIV-1 agents for training the neural network, to make  molecular docking of all compounds from this library with gp120, to  calculate the values of binding free energy, to generate molecular fingerprints for chemical compounds from the training dataset. The training the neural network was implemented followed by estimation of the learning outcomes and work of the autoencoder.  The validation of the neural network on a wide range of compounds from the ZINC database was carried out. The use of the neural network in combination with virtual screening of chemical databases was shown to form a productive platform for identifying the basic structures promising for the design of novel antiviral drugs that inhibit the early stages of HIV infection.

Friction Compensation in the Inverted Pendulum Controller by Means of a Neural Network

2020 ◽  
Vol 22 (4) ◽  
pp. 875-884
Author(s):  
Marek Balcerzak
Keyword(s):  
Neural Network ◽  
Inverted Pendulum ◽  
Friction Model ◽  
The Novel ◽  
Practical Procedure ◽  
The Neural Network ◽  
Wide Range ◽  
Novel Method ◽  
Data Acquisition And Processing ◽  
Friction Modelling

AbstractThis paper presents an experimental confirmation of the novel method of friction modelling and compensation. The method has been applied to an inverted pendulum control system. The practical procedure of data acquisition and processing has been described. Training of the neural network friction model has been covered. Application of the obtained model has been presented. The main asset of the presented model is its correctness in a wide range of relative velocities. Moreover, the model is relatively easy to build.

Vehicle Sideslip Angle Estimation Through Neural Networks: Application to Experimental Data

2006 ◽  
Author(s):  
Stefano Melzi ◽  
Edoardo Sabbioni ◽  
Alessandro Concas ◽  
Marco Pesce
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Lateral Acceleration ◽  
Slip Angle ◽  
Vehicle Speed ◽  
Sideslip Angle ◽  
The Neural Network ◽  
Training Stage ◽  
Wide Range ◽  
Self Adaptation

This work explores the possibility of using a non-structured algorithm as a sideslip angle valuer: on the basis of a preliminary numerical analysis, a neural network was designed and trained with experimental signals of lateral acceleration, vehicle speed, yaw rate and steer angle. The network was applied to experimental data in order to verify its capability of self-adaptation to changes in friction coefficient and to provide accurate estimations for manoeuvres sensibly different from the ones used during the training stage. The simple architecture joined with an appropriate training set conferred good self-adaptation properties to the neural network which was able to provide satisfying estimation of side slip angle for a wide range of manoeuvres and different friction conditions.

scholarly journals Classification of Plant Leaf Diseases Based on Improved Convolutional Neural Network

Sensors ◽  
2019 ◽  
Vol 19 (19) ◽  
pp. 4161 ◽  
Author(s):  
Hang ◽  
Zhang ◽  
Chen ◽  
Zhang ◽  
Wang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Convergence Time ◽  
Model Parameters ◽  
Data Set ◽  
Training Time ◽  
Plant Leaf ◽  
The Neural Network ◽  
Target Disease ◽  
Fully Connected

Plant leaf diseases are closely related to people's daily life. Due to the wide variety of diseases, it is not only time-consuming and labor-intensive to identify and classify diseases by artificial eyes, but also easy to be misidentified with having a high error rate. Therefore, we proposed a deep learning-based method to identify and classify plant leaf diseases. The proposed method can take the advantages of the neural network to extract the characteristics of diseased parts, and thus to classify target disease areas. To address the issues of long training convergence time and too-large model parameters, the traditional convolutional neural network was improved by combining a structure of inception module, a squeeze-and-excitation (SE) module and a global pooling layer to identify diseases. Through the Inception structure, the feature data of the convolutional layer were fused in multi-scales to improve the accuracy on the leaf disease dataset. Finally, the global average pooling layer was used instead of the fully connected layer to reduce the number of model parameters. Compared with some traditional convolutional neural networks, our model yielded better performance and achieved an accuracy of 91.7% on the test data set. At the same time, the number of model parameters and training time have also been greatly reduced. The experimental classification on plant leaf diseases indicated that our method is feasible and effective.

Optimization of neural network parameters using Grey–Taguchi methodology for manufacturing process applications

2014 ◽  
Vol 229 (14) ◽  
pp. 2651-2664 ◽  
Author(s):  
Dinesh Kumar ◽  
Arun Kumar Gupta ◽  
Pankaj Chandna ◽  
Mahesh Pal
Keyword(s):  
Neural Network ◽  
Depth Of Cut ◽  
Taguchi Methodology ◽  
Training Time ◽  
Network Parameters ◽  
The Neural Network ◽  
Trial And Error Method ◽  
Grey Relational ◽  
Grey Taguchi ◽  
Error Method

Performance of neural networks depends upon several input parameters. Several attempts have been made for optimization of neural network parameters using Taguchi methodology for achieving single objective such as computation effort, computation time, etc. Determination of optimum setting to these parameters still remains a difficult task. Trial-and-error method is one of the frequently used approaches to determine the optimal choice of these parameters. Keeping in view the problems with trial-and-error method, a systematic approach is required to find the optimum value of different parameters of neural network. In the present work, three most important distinct performance measures such as mean square error between actual and prediction, number of iteration, and total training time consumption have been probably considered first time concurrently. The multiobjective problem has been solved using Grey–Taguchi methodology. In this study, optimal combinations of different neural network parameters have been identified by using the Taguchi-based Grey relational analysis. The data set includes 81 sets of milling data corresponding to three-level full factorial experimental design for four cutting parameters, i.e. cutting speed, feed, axial depth of cut, and radial depth of cut, respectively. The output is average surface roughness for the experiment. The performance of different neural network models has been tabulated in L36 orthogonal array. Confidence interval has also been estimated for 95% consistency level to validate the optimum level of different parameters. It was found that the Taguchi-based Grey relational analysis approach can effectively be used as a structured method to optimize the neural network parameters settings, which can be easily implemented to enhance the performance of the neural network model with a relatively small size and time saving experiment. The result clearly indicates that the optimal combination of neural network parameters obtained by using the proposed approach performs better in terms of low mean square error, small number of iterations, and lesser training time required to perform the analysis which further results in lesser computation effort and processing time. Methodology proposed in this work can be utilized for any type of neural network application to find the optimum levels of different parameters.

Sign in / Sign up

Export Citation Format

Share Document