scholarly journals Analyzing the Effect of Hyper-Parameters on a Neural Network for Emotion Detection

2021 ◽  
pp. 33-40
Author(s):  
Ambrish Jhamnani ◽  
Anshika Tiwari ◽  
Abhishek Soni ◽  
Arpit Deo
Keyword(s):  
Neural Network ◽  
Prediction Model ◽  
Optimal Solution ◽  
Poor Performance ◽  
Activation Function ◽  
Human Face ◽  
Emotion Detection ◽  
Good Balance ◽  
Human Emotion ◽  
The Poor

Human emotion prediction is a tough task. The human face is extremely complex to understand. To build an optimal solution for human emotion prediction model, setting hyper-parameter plays a major role. It is a difficult task to train a neural network. The poor performance of the model can result from poor judgment of sub-optimal hyper- parameters before training the model. This study aims to compare different hyper-parameters and their effect to train the convolutional neural network for emotion detection. We used different methods based on values of validation accuracy and validation loss. The study reveals that SELU activation function performs better in terms of validation accuracy. Swish activation function maintains a good balance between validation accuracy and validation loss. As different combinations of parameters behave differently likewise in optimizers, RMS prop gives less validation loss with Swish whereas Adam performs better with ReLU and ELU activation function.

A One-Layer Recurrent Neural Network with a Discontinuous Activation Function for Linear Programming

2008 ◽  
Vol 20 (5) ◽  
pp. 1366-1383 ◽  
Author(s):  
Qingshan Liu ◽  
Jun Wang
Keyword(s):  
Neural Network ◽  
Linear Programming ◽  
Recurrent Neural Network ◽  
Optimal Solution ◽  
Activation Function ◽  
High Gain ◽  
Discontinuous Activation ◽  
The Neural Network ◽  
Decision Variables ◽  
Discontinuous Activation Function

A one-layer recurrent neural network with a discontinuous activation function is proposed for linear programming. The number of neurons in the neural network is equal to that of decision variables in the linear programming problem. It is proven that the neural network with a sufficiently high gain is globally convergent to the optimal solution. Its application to linear assignment is discussed to demonstrate the utility of the neural network. Several simulation examples are given to show the effectiveness and characteristics of the neural network.

The Prediction Model of the Mechanical Properties of Composite Materials Based on Ant Colony Neural Network

2013 ◽  
Vol 680 ◽  
pp. 39-43
Author(s):  
Jing Wang ◽  
Jie Zhu ◽  
Qian Zhang
Keyword(s):  
Neural Network ◽  
Mechanical Properties ◽  
Composite Materials ◽  
Prediction Model ◽  
Ant Colony Algorithm ◽  
Optimal Solution ◽  
Ant Colony ◽  
Practical Applications ◽  
Global Optimal ◽  
Adaptive Ability

In this paper, a prediction model of the mechanical properties of composite materials has been proposed based on the ant colony neural network. The mechanical properties of the materials are the common problems that the various materials must be involved in the practical applications. The testing of the mechanical properties of the composite materials is of great significance to the development and the progress of the theory and the practice of composite materials. The ant colony algorithm takes advantage of the optimization mechanisms of ant colony, which has a strong ability to find the global optimal solution. The candidate group mechanism is added in the ant colony algorithm and the weights of the artificial neural network are trained through using the improved ant colony algorithm. This model has a strong adaptive ability and can be used in the prediction of the mechanical properties of composite materials. Then, the efficiency of the testing of mechanical properties can be improved.

scholarly journals Prediction of springback in local bending of hull plates using an optimized backpropagation neural network

2021 ◽  
Vol 12 (2) ◽  
pp. 777-789
Author(s):  
Binjiang Xu ◽  
Lei Li ◽  
Zhao Wang ◽  
Honggen Zhou ◽  
Di Liu
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Prediction Model ◽  
Optimal Solution ◽  
Backpropagation Neural Network ◽  
Processing Efficiency ◽  
Model Based ◽  
Assembly Accuracy ◽  
Bending Process ◽  
Springback Prediction

Abstract. Springback is an inevitable problem in the local bending process of hull plates, which leads to low processing efficiency and affects the assembly accuracy. Therefore, the prediction of the springback effect, as a result of the local bending of hull plates, bears great significance. This paper proposes a springback prediction model based on a backpropagation neural network (BPNN), considering geometric and process parameters. Genetic algorithm (GA) and improved particle swarm optimization (PSO) algorithms are used to improve the global search capability of BPNN, which tends to fall into local optimal solutions, in order to find the global optimal solution. The result shows that the proposed springback prediction model, based on the BPNN optimized by genetic algorithm, is faster and offers smaller prediction error on the springback due to local bending.

scholarly journals GA-BP Neural Network Regression Model for Predicting the Soil Moisture of Ecological Slope Protection

2021 ◽  
Author(s):  
Dunwen Liu ◽  
Chao Liu ◽  
Yu Tang ◽  
Chun Gong
Keyword(s):  
Neural Network ◽  
Soil Moisture ◽  
Prediction Model ◽  
High Speed ◽  
Meteorological Factors ◽  
Meteorological Data ◽  
Prediction Method ◽  
Optimal Solution ◽  
Percentage Error ◽  
The Neural Network

Abstract The neural network optimized by genetic algorithm(GA) is an efficient and accurate prediction method, which can quickly find the optimal solution through high-speed computing capability and self-learning function. The neural network model optimized by GA is applied to the prediction of soil moisture of ecological slope protection, which provides reference for practical application of slope vegetation screening. In this paper, nine meteorological factors and soil moisture data were obtained by field monitoring instruments and related meteorological data. Considering the lag of meteorological factors, the neural network optimized by GA is used to predict the soil moisture of 8 meteorological data. The results show that the root mean square error (RMSE) and the mean absolute percentage error (MAPE) of the prediction model are 0.22726 and 0.41234%, respectively, indicating that the prediction model runs well. Through V-fold cross-validation, it is found that the prediction results of the model is accurate and stable. The algorithm combining artificial neural network and GA can well predict the soil moisture of ecological slope protection, with high prediction accuracy, and has a good application prospect in other fields.

Stock Market Prediction using Data Mining Techniques

2020 ◽  
pp. 310-313
Author(s):  
Karunesh Makker ◽  
Prince Patel ◽  
Hrishikesh Roy ◽  
Sonali Borse
Keyword(s):  
Neural Network ◽  
Stock Market ◽  
Prediction Model ◽  
Efficient Market Hypothesis ◽  
Deterministic System ◽  
Vast Number ◽  
Factors Influencing ◽  
Number Of Factors ◽  
Using Data ◽  
Robust Prediction

Stock market is a very volatile in-deterministic system with vast number of factors influencing the direction of trend on varying scales and multiple layers. Efficient Market Hypothesis (EMH) states that the market is unbeatable. This makes predicting the uptrend or downtrend a very challenging task. This research aims to combine multiple existing techniques into a much more robust prediction model which can handle various scenarios in which investment can be beneficial. Existing techniques like sentiment analysis or neural network techniques can be too narrow in their approach and can lead to erroneous outcomes for varying scenarios. By combing both techniques, this prediction model can provide more accurate and flexible recommendations. Embedding Technical indicators will guide the investor to minimize the risk and reap better returns.

SCORING MODELING BASED ON NEURAL NETWORKS FOR DETERMINING A BANK BORROWER'S RATING

2020 ◽  
Vol 2020 (10) ◽  
pp. 54-62
Author(s):  
Oleksii VASYLIEV ◽  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Architecture ◽  
Statistical Data ◽  
Activation Function ◽  
Decision Making Process ◽  
Neural Network Architecture ◽  
Acceptable Accuracy ◽  
The Neural Network ◽  
Sigmoid Activation Function

The problem of applying neural networks to calculate ratings used in banking in the decision-making process on granting or not granting loans to borrowers is considered. The task is to determine the rating function of the borrower based on a set of statistical data on the effectiveness of loans provided by the bank. When constructing a regression model to calculate the rating function, it is necessary to know its general form. If so, the task is to calculate the parameters that are included in the expression for the rating function. In contrast to this approach, in the case of using neural networks, there is no need to specify the general form for the rating function. Instead, certain neural network architecture is chosen and parameters are calculated for it on the basis of statistical data. Importantly, the same neural network architecture can be used to process different sets of statistical data. The disadvantages of using neural networks include the need to calculate a large number of parameters. There is also no universal algorithm that would determine the optimal neural network architecture. As an example of the use of neural networks to determine the borrower's rating, a model system is considered, in which the borrower's rating is determined by a known non-analytical rating function. A neural network with two inner layers, which contain, respectively, three and two neurons and have a sigmoid activation function, is used for modeling. It is shown that the use of the neural network allows restoring the borrower's rating function with quite acceptable accuracy.

The Roma in Post-Communist Bulgaria: Growing Social Marginalization and State Policies

2020 ◽  
Vol 2 (1) ◽  
pp. 1-24
Author(s):  
Yorgos Christidis
Keyword(s):  
Political Parties ◽  
Social Exclusion ◽  
Hate Crime ◽  
Hate Speech ◽  
Poor Performance ◽  
State Policies ◽  
Social Marginalization ◽  
Popular Support ◽  
Economic Problems ◽  
The Poor

This article analyzes the growing impoverishment and marginalization of the Roma in Bulgarian society and the evolution of Bulgaria’s post-1989 policies towards the Roma. It examines the results of the policies so far and the reasons behind the “poor performance” of the policies implemented. It is believed that Post-communist Bulgaria has successfully re-integrated the ethnic Turkish minority given both the assimilation campaign carried out against it in the 1980s and the tragic events that took place in ex-Yugoslavia in the 1990s. This Bulgaria’s successful “ethnic model”, however, has failed to include the Roma. The “Roma issue” has emerged as one of the most serious and intractable ones facing Bulgaria since 1990. A growing part of its population has been living in circumstances of poverty and marginalization that seem only to deteriorate as years go by. State policies that have been introduced since 1999 have failed at large to produce tangible results and to reverse the socio-economic marginalization of the Roma: discrimination, poverty, and social exclusion continue to be the norm. NGOs point out to the fact that many of the measures that have been announced have not been properly implemented, and that legislation existing to tackle discrimination, hate crime, and hate speech is not implemented. Bulgaria’s political parties are averse in dealing with the Roma issue. Policies addressing the socio-economic problems of the Roma, including hate speech and crime, do not enjoy popular support and are seen as politically damaging.

Sign in / Sign up

Export Citation Format

Share Document