Open Access. Powered by Scholars. Published by Universities.®
Operations Research, Systems Engineering and Industrial Engineering Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 9 of 9
Full-Text Articles in Operations Research, Systems Engineering and Industrial Engineering
Forecasting Series-Based Stock Price Data Using Direct Reinforcement Learning, H. Li, Cihan H. Dagli, David Lee Enke
Forecasting Series-Based Stock Price Data Using Direct Reinforcement Learning, H. Li, Cihan H. Dagli, David Lee Enke
Engineering Management and Systems Engineering Faculty Research & Creative Works
A significant amount of work has been done in the area of price series forecasting using soft computing techniques, most of which are based upon supervised learning. Unfortunately, there has been evidence that such models suffer from fundamental drawbacks. Given that the short-term performance of the financial forecasting architecture can be immediately measured, it is possible to integrate reinforcement learning into such applications. In this paper, we present the novel hybrid view for a financial series and critic adaptation stock price forecasting architecture using direct reinforcement. A new utility function called policies-matching ratio is also proposed. The need for the …
Simple Ensemble-Averaging Model Based On Generalized Regression Neural Network In Financial Forecasting Problems, Cihan H. Dagli, Parinya Disorntetiwat
Simple Ensemble-Averaging Model Based On Generalized Regression Neural Network In Financial Forecasting Problems, Cihan H. Dagli, Parinya Disorntetiwat
Engineering Management and Systems Engineering Faculty Research & Creative Works
Introduces an ensemble-averaging model based on a GRNN (generalized regression neural network) for financial forecasting. The model trains all input individually using GRNNs and uses a simple ensemble-averaging committee machine to improve the accuracy performance. In a financial problem, there are many different factors that can effect the asset price movement at different times. An experiment is implemented in two different data sets, S&P 500 index and currency exchange rate. The predictive abilities of the model are evaluated on the basis of root mean squared error, standard deviation and percent direction correctness. The study shows a promising result of the …
A Biologically Inspired Connectionist Model For Image Feature Extraction In 2d Pattern Recognition, Cihan H. Dagli, Raymond K. Chafin
A Biologically Inspired Connectionist Model For Image Feature Extraction In 2d Pattern Recognition, Cihan H. Dagli, Raymond K. Chafin
Engineering Management and Systems Engineering Faculty Research & Creative Works
A new edge detection method is presented which borrows from recent research into primate vision biology, and offers improved noise performance over classical methods. Beginning with spatio-temporal shunting models for retinal cones, horizontal cells, bipolar cells, and retinal ganglions, a set of simplified steady-state solutions are developed which lend themselves to efficient computation on standard computer equipment. The retinal model output is found to be nominally equivalent to the classical edge detector, but is produced differently. A simplified model of the lateral geniculate nucleus (LGN) has been produced. Taking the output of the retinal model, the LGN simple cell and …
A Hybrid System For Well Test Analysis, Edward A. May, Cihan H. Dagli
A Hybrid System For Well Test Analysis, Edward A. May, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
Petroleum well test analysis is a tool for estimating the average properties of the reservoir rock. It is a classic example of an inverse problem. Visual examination of the pressure response of the reservoir to an induced flow rate change at a well allows the experienced analyst to determine the most appropriate model from a library of generalized analytical solutions. Rock properties are determined by finding the model parameters that best fit the observed data. This paper describes a framework for hybrid network to assist the analyst in selecting the appropriate model and determining the solution. The hybrid network design …
A Comparison Of Fam And Cmac For Nonlinear Control, Arit Thammano, Cihan H. Dagli
A Comparison Of Fam And Cmac For Nonlinear Control, Arit Thammano, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
This article compares a neural network-based controller, both local and global networks, with fuzzy associative memories (FAM) on a nonlinear problem. CMAC and FAM are chosen as representatives of local generalization networks. CMAC controller is trained off-line, therefore, it can response to the incoming input immediately. CMAC can interpolate its memory and give a reasonable control signal even the input has not been trained on. Backpropagation is picked as a representative of global generalization networks. All three systems are studied on a simple simulated control problem. This preliminary research will be adapted later to control the laser cutting machine. A …
Integrated Neural Network And Machine Vision Approach For Intelligent State Identification, Cihan H. Dagli, Timothy Andrew Bauer
Integrated Neural Network And Machine Vision Approach For Intelligent State Identification, Cihan H. Dagli, Timothy Andrew Bauer
Engineering Management and Systems Engineering Faculty Research & Creative Works
An interfacing of neural networks (NNs) and machine vision to provide the next state of a system given an image of the present state of the system is presented. This interfacing is applied to a loading operation. First, a NN is trained for part recognition under conditions of rotation, location, object distortion, and background noise given an image of the part. Then, a second NN, given the output of the first NN and an image of a pallet being loaded, is trained for optimal part loading onto the pallet under conditions of noise in the image. The paradigm used is …
An Empirical Analysis Of Backpropagation Error Surface Initiation For Injection Molding Process Control, Alice E. Smith, Elaine R. Raterman, Cihan H. Dagli
An Empirical Analysis Of Backpropagation Error Surface Initiation For Injection Molding Process Control, Alice E. Smith, Elaine R. Raterman, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
Backpropagation neural networks are trained by adjusting initially random interconnecting weights according to the steepest local error surface gradient. The authors examine the practical implications of the arbitrary starting point on the error landscape of the ensuing trained network. The effects on network convergence and performance are tested empirically, varying parameters such as network size, training rate, transfer function and data representation. The data used are live process control data from an injection molding plant
Neural Networks In Manufacturing: Possible Impacts On Cutting Stock Problems, Cihan H. Dagli
Neural Networks In Manufacturing: Possible Impacts On Cutting Stock Problems, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
The potential of neural networks is examined, and the effect of parallel processing on the solution of the stock-cutting problem is assessed. The conceptual model proposed integrates a feature-recognition network and a simulated annealing approach. The model uses a neocognitron neural network paradigm to generate data for assessing the degree of match between two irregular patterns. The information generated through the feature recognition network is passed to an energy function, and the optimal configuration of patterns is computed using a simulated annealing algorithm. Basics of the approach are demonstrated with an example.
Possible Applications Of Neural Networks In Manufacturing, S. Lammers, Cihan H. Dagli
Possible Applications Of Neural Networks In Manufacturing, S. Lammers, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
Summary form only given. An examination is made of the potential of neural networks and the impact of parallel processing in the design and operations of manufacturing systems. After an initial discussion on possible areas of application, an approach that integrates artificial intelligence, operations research, and neural networks for the solution of a scheduling problem is examined