scholarly journals Verification of RNN-Based Neural Agent-Environment Systems

2019 ◽  
Vol 33 ◽  
pp. 6006-6013 ◽  
Author(s):  
Michael E. Akintunde ◽  
Andreea Kevorchian ◽  
Alessio Lomuscio ◽  
Edoardo Pirovano
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Recurrent Neural Network ◽  
Experimental Results ◽  
Feed Forward ◽  
Execution Traces ◽  
Verification Problem ◽  
Complete Procedure ◽  
Feed Forward Neural Networks

We introduce agent-environment systems where the agent is stateful and executing a ReLU recurrent neural network. We define and study their verification problem by providing equivalences of recurrent and feed-forward neural networks on bounded execution traces. We give a sound and complete procedure for their verification against properties specified in a simplified version of LTL on bounded executions. We present an implementation and discuss the experimental results obtained.

Missing Data Approaches to Classification

2010 ◽  
pp. 187-209
Author(s):  
Tshilidzi Marwala
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Missing Data ◽  
Network Model ◽  
Network Models ◽  
Feed Forward Neural Network ◽  
Neural Network Models ◽  
Feed Forward ◽  
Missing Data Problem ◽  
Feed Forward Neural Networks

In this chapter, a classifier technique that is based on a missing data estimation framework that uses autoassociative multi-layer perceptron neural networks and genetic algorithms is proposed. The proposed method is tested on a set of demographic properties of individuals obtained from the South African antenatal survey and compared to conventional feed-forward neural networks. The missing data approach based on the autoassociative network model proposed gives an accuracy of 92%, when compared to the accuracy of 84% obtained from the conventional feed-forward neural network models. The area under the receiver operating characteristics curve for the proposed autoassociative network model is 0.86 compared to 0.80 for the conventional feed-forward neural network model. The autoassociative network model proposed in this chapter, therefore, outperforms the conventional feed-forward neural network models and is an improved classifier. The reasons for this are: (1) the propagation of errors in the autoassociative network model is more distributed while for a conventional feed-forward network is more concentrated; and (2) there is no causality between the demographic properties and the HIV and, therefore, the HIV status does change the demographic properties and vice versa. Therefore, it is better to treat the problem as a missing data problem rather than a feed-forward problem.

scholarly journals Verification of Closed-loop Systems with Neural Network Controllers

10.29007/btv1 ◽  
2019 ◽  
Author(s):  
Diego Manzanas Lopez ◽  
Patrick Musau ◽  
Hoang-Dung Tran ◽  
Taylor T. Johnson
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Control Systems ◽  
Closed Loop ◽  
Closed Loop Control ◽  
Feed Forward ◽  
Loop Control ◽  
Closed Loop Systems ◽  
Feed Forward Neural Networks ◽  
Neural Network Controllers

This benchmark suite presents a detailed description of a series of closed-loop control systems with artificial neural network controllers. In many applications, feed-forward neural networks are heavily involved in the implementation of controllers by learning and representing control laws through several methods such as model predictive control (MPC) and reinforcement learning (RL). The type of networks that we consider in this manuscript are feed-forward neural networks consisting of multiple hidden layers with ReLU activation functions and a linear activation function in the output layer. While neural network con- trollers have been able to achieve desirable performance in many contexts, they also present a unique challenge in that it is difficult to provide any guarantees about the correctness of their behavior or reason about the stability a system that employs their use. Thus, from a controls perspective, it is necessary to verify them in conjunction with their corresponding plants in closed-loop. While there have been a handful of works proposed towards the verification of closed-loop systems with feed-forward neural network controllers, this area still lacks attention and a unified set of benchmark examples on which verification techniques can be evaluated and compared. Thus, to this end, we present a range of closed-loop control systems ranging from two to six state variables, and a range of controllers with sizes in the range of eleven neurons to a few hundred neurons in more complex systems.

scholarly journals Formal verification of neural agents in non-deterministic environments

2021 ◽  
Vol 36 (1) ◽  
Author(s):  
Michael E. Akintunde ◽  
Elena Botoeva ◽  
Panagiotis Kouvaros ◽  
Alessio Lomuscio
Keyword(s):  
Neural Networks ◽  
Parallel Algorithms ◽  
Formal Verification ◽  
Experimental Results ◽  
Use Case ◽  
Feed Forward ◽  
Verification Problem ◽  
Bounded Fragment

AbstractWe introduce a model for agent-environment systems where the agents are implemented via feed-forward ReLU neural networks and the environment is non-deterministic. We study the verification problem of such systems against CTL properties. We show that verifying these systems against reachability properties is undecidable. We introduce a bounded fragment of CTL, show its usefulness in identifying shallow bugs in the system, and prove that the verification problem against specifications in bounded CTL is in coNExpTime and PSpace-hard. We introduce sequential and parallel algorithms for MILP-based verification of agent-environment systems, present an implementation, and report the experimental results obtained against a variant of the VerticalCAS use-case and the frozen lake scenario.

Estimation of lateral-directional parameters using neural networks based modified delta method

2007 ◽  
Vol 111 (1124) ◽  
pp. 659-667 ◽  
Author(s):  
S. Singh ◽  
A. K. Ghosh
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Flight Control ◽  
Delta Method ◽  
Feed Forward ◽  
Flight Data ◽  
Stability And Control ◽  
Control Derivatives ◽  
Feed Forward Neural Networks ◽  
And Control

Abstract The aim of the study described herein was to develop and verify an efficient neural network based method for extracting aircraft stability and control derivatives from real flight data using feed-forward neural networks. The proposed method (Modified Delta method) draws its inspiration from feed forward neural network based the Delta method for estimating stability and control derivatives. The neural network is trained using differential variation of aircraft motion/control variables and coefficients as the network inputs and outputs respectively. For the purpose of parameter estimation, the trained neural network is presented with a suitably modified input file and the corresponding predicted output file of aerodynamic coefficients is obtained. An appropriate interpretation and manipulation of such input-output files yields the estimates of the parameter. The method is validated first on the simulated flight data using various combinations and types of real-flight control inputs and then on real flight data. A new technique is also proposed for validating the estimated parameters using feed-forward neural networks.

scholarly journals A Neural Network Approximation Based on a Parametric Sigmoidal Function

Mathematics ◽  
2019 ◽  
Vol 7 (3) ◽  
pp. 262 ◽  
Author(s):  
Beong Yun
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Approximation Method ◽  
Closed Interval ◽  
Activation Function ◽  
Sigmoidal Function ◽  
Feed Forward Neural Network ◽  
Feed Forward ◽  
Extended Application ◽  
Feed Forward Neural Networks

It is well known that feed-forward neural networks can be used for approximation to functions based on an appropriate activation function. In this paper, employing a new sigmoidal function with a parameter for an activation function, we consider a constructive feed-forward neural network approximation on a closed interval. The developed approximation method takes a simple form of a superposition of the parametric sigmoidal function. It is shown that the proposed method is very effective in approximation of discontinuous functions as well as continuous ones. For some examples, the availability of the presented method is demonstrated by comparing its numerical results with those of an existing neural network approximation method. Furthermore, the efficiency of the method in extended application to the multivariate function is also illustrated.

scholarly journals Algorithm for calculating synapse weights of the first layer of a neural network on the base of metric recognition methods. Part 1.

2020 ◽  
pp. 20-30
Author(s):  
Polad Geidarov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Three Dimensional ◽  
Analytical Calculation ◽  
Training Algorithms ◽  
Feed Forward Neural Network ◽  
Software Environment ◽  
Feed Forward ◽  
Feed Forward Neural Networks ◽  
And Training

Introduction: Metric recognition methods make it possible to preliminarily and strictly determine the structures of feed-forward neural networks, namely, the number of neurons, layers, and connections based on the initial parameters of the recognition problem. They also make it possible to analytically calculate the synapse weights of network neurons based on metric expressions. The setup procedure for these networks includes a sequential analytical calculation of the values of each synapse weight in the weight table for neurons of the zero or first layer, which allows us to obtain a working feed-forward neural network at the initial stage without the use of training algorithms. Then feed-forward neural networks can be trained by well-known learning algorithms, which generally speeds up the process of their creation and training. Purpose: To determine how much time the process of calculating the values of weights requires and, accordingly, how reasonable it is to preliminarily calculate the weights of a feed-forward neural network. Results: An algorithm is proposed and implemented for the automated calculation of all values of synapse weight tables for the zero and first layers as applied to the task of recognizing black-and-white monochrome symbol images. The proposed algorithm is described in the Builder C++ software environment. The possibility of optimizing the process of calculating the weights of synapses in order to accelerate the entire algorithm is considered. The time spent on calculating these weights for different configurations of neural networks based on metric recognition methods is estimated. Examples of creating and calculating synapse weight tables according to the considered algorithm are given. According to them, the analytical calculation of the weights of a neural network takes just seconds or minutes, being in no way comparable to the time necessary for training a neural network. Practical relevance: Analytical calculation of the weights of a neural network can significantly accelerate the process of creating and training a feed-forward neural network. Based on the proposed algorithm, we can implement one for calculating three-dimensional weight tables for more complex images, either blackand-white or color grayscale ones.

scholarly journals Algorithm for calculating synapse weights of the first layer of a neural network on the base of metric recognition methods. Part 2

2020 ◽  
pp. 25-38
Author(s):  
Polad Geidarov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Analytical Calculation ◽  
Training Algorithms ◽  
Feed Forward Neural Network ◽  
Software Environment ◽  
Feed Forward ◽  
Black And White ◽  
Feed Forward Neural Networks ◽  
And Training

Introduction: Metric recognition methods make it possible to preliminarily and strictly determine the structures of feed-forward neural networks, namely, the number of neurons, layers, and connections based on the initial parameters of the recognition problem. They also make it possible to analytically calculate the synapse weights of network neurons based on metric expressions. The setup procedure for these networks includes a sequential analytical calculation of the values of each synapse weight in the weight table for neurons of the zero or first layer, which allows us to obtain a working feed-forward neural network at the initial stage without the use of training algorithms. Then feed-forward neural networks can be trained by well-known learning algorithms, which generally speeds up the process of their creation and training. Purpose: To determine how much time the process of calculating the values of weights requires and, accordingly, how reasonable it is to preliminarily calculate the weights of a feed-forward neural network. Results: An algorithm is proposed and implemented for the automated calculation of all values of synapse weight tables for the zero and first layers as applied to the task of recognizing black-and-white monochrome symbol images. The proposed algorithm is described in the Builder C++ software environment. The possibility of optimizing the process of calculating the weights of synapses in order to accelerate the entire algorithm is considered. The time spent on calculating these weights for different configurations of neural networks based on metric recognition methods is estimated. Examples of creating and calculating synapse weight tables according to the considered algorithm are given. According to them, the analytical calculation of the weights of a neural network takes just seconds or minutes, being in no way comparable to the time necessary for training a neural network. Practical relevance: Analytical calculation of the weights of a neural network can significantly accelerate the process of creating and training a feed-forward neural network. Based on the proposed algorithm, we can implement one for calculating three-dimensional weight tables for more complex images, either black and-white or color grayscale ones.

A Novel Composite Neural Network for Hysteresis Modeling in Piezoelectric Actuators

2009 ◽  
Author(s):  
Mohamad Fazli ◽  
Seyed Mahdi Rezaei ◽  
Mohamad Zareienejad
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Piezoelectric Actuators ◽  
Maximum Error ◽  
Feed Forward ◽  
Positioning Systems ◽  
Time Dynamic ◽  
Operational Conditions ◽  
Hysteresis Modeling ◽  
Feed Forward Neural Networks

Piezoelectric actuators are convenient for micro positioning systems. Inherent hysteresis is one of the drawbacks in use of these actuators. Precise control of this actuator under changing of environmental and operational conditions, without modeling of hysteresis, is impossible. Neural networks can be used for this modeling. The ordinary feed forward neural networks can not train time dynamic relationship between input and output. Thus a certain type of networks called time delay feed forward neural networks (TDNN), are developed and is used in this paper. In the previous researches in this field, the important effect of loaded force on the actuator is ignored. This can increase the positioning error remarkably. Especially when these actuators are used in the precise grinding or machining operations. In this paper, neural network is used for hysteresis modeling with attention to the important effect of loaded force. After modeling, inverse hysteresis model is used as compensator in a feed forward way to linearize the input-output relationship. Then using PI closed loop controller and selecting suitable coefficient for it, the maximum error was decreased to less than 2 percent of the working amplitude.

scholarly journals Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations

2004 ◽  
Vol 4 (3) ◽  
pp. 3653-3667 ◽  
Author(s):  
D. J. Lary ◽  
H. Y. Mussa
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Kalman Filter ◽  
Extended Kalman Filter ◽  
Learning Algorithm ◽  
Feed Forward ◽  
The Neural Network ◽  
Fortran Code ◽  
Feed Forward Neural Networks ◽  
Time Of Year

Abstract. In this study a new extended Kalman filter (EKF) learning algorithm for feed-forward neural networks (FFN) is used. With the EKF approach, the training of the FFN can be seen as state estimation for a non-linear stationary process. The EKF method gives excellent convergence performances provided that there is enough computer core memory and that the machine precision is high. Neural networks are ideally suited to describe the spatial and temporal dependence of tracer-tracer correlations. The neural network performs well even in regions where the correlations are less compact and normally a family of correlation curves would be required. For example, the CH4-N2O correlation can be well described using a neural network trained with the latitude, pressure, time of year, and CH4 volume mixing ratio (v.m.r.). The neural network was able to reproduce the CH4-N2O correlation with a correlation coefficient between simulated and training values of 0.9997. The neural network Fortran code used is available for download.

scholarly journals Using A Feed Forward Neural Network Algorithm to Predict Prices of Multiple Cryptocurrencies

2021 ◽  
Vol 6 (5) ◽  
pp. 15-19
Author(s):  
Sina E. Charandabi ◽  
Kamyar Kamyar
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Learning Model ◽  
Feed Forward Neural Network ◽  
Feed Forward ◽  
Price Data ◽  
Machine Learning Model ◽  
Feed Forward Neural Networks ◽  
Neural Network Algorithm

This paper initially presents a nontechnical overview of cryptocurrency, its history, and the technicalities of its usage as a means of exchange. Bitcoin’s working methodology and mathematical baseline is further presented in more depth. For the remaining majority of the paper, recent cryptocurrency price data of Bitcoin, Ethereum, Tether, Dogecoin, and Binance coin was used to train a machine learning model of Feed Forward Neural Networks to predict future prices for each of the datasets. Further and in conclusion, the results are discussed, and the efficiency and accuracy of these models are evaluated.

Sign in / Sign up

Export Citation Format

Share Document