Extension Of Genetic Programming Research Articles

Efficiency improvement of genetic network programming by tasks decomposition in different types of environments

Genetic Network Programming (GNP) is a relatively recently proposed evolutionary algorithm which is an extension of Genetic Programming (GP). However, individuals in GNP have graph structures. This algorithm is mainly used in decision making process of agent control problems. It uses a graph to make a flowchart and use this flowchart as a decision making strategy that an agent must follow to achieve the goal. One of the most important weaknesses of this algorithm is that crossover and mutation break the structures of individuals during the evolution process. Although it can lead to better structures, this may break suitable ones and increase the time needed to achieve optimal solutions. Meanwhile, all the researches in this field are dedicated to test GNP in deterministic environments. However, most of the real-world problems are stochastic and this is another issue that should be addressed. In this research, we try to find a mechanism that GNP shows better performance in stochastic environments. In order to achieve this goal, the evolution process of GNP was modified. In the proposed method, the experience of promising individuals was saved in consecutive generations. Then, to generate offspring in some predefined number of generations, the saved experiences were used instead of crossover and mutation. The experimental results of the proposed method were compared with GNP and some of its versions in both deterministic and stochastic environments. The results demonstrate the superiority of our proposed method in both deterministic and stochastic environments.

A laboratory investigation on the potential of computational intelligence approaches to estimate the discharge coefficient of piano key weir

The piano key weir (PKW) is a type of nonlinear control structure that can be used to increase unit discharge over linear overflow weir geometries, particularly when the weir footprint area is restricted To predict the outflow passing over a piano key weir, the discharge coefficient in the general equation of weir needs to be known. This paper presents the results of laboratory model testing of a piano key weir located on the straight open channel flume in the hydraulic laboratory of Bu-Ali Sina University. The discharge coefficient of piano key weir is estimated by using four computational intelligence approaches, namely, feed forward back-propagation neural network (FFBPN), an extension of genetic programming namely gene-expression programming (GEP), least square support vector machine (LSSVM) and extreme learning machine (ELM). For this purpose, 70 laboratory test results were used for determining discharge coefficient of piano key weir for a wide range of discharge values. Coefficient of determination (R2), Nash-Sutcliffe efficiency coefficient (NS), root mean square error (RMSE), mean absolute relative error (MARE), scatter index (SI) and BIAS are used for measuring the models’ performance. Overall performance of the models shows that, all the studied models are able to estimate discharge coefficient of piano key weir satisfactorily. Comparison of results showed that the ELM (R2=0.997 and NS= 0.986) and LSSVM (RMSE=0.016 and MARE=0.027) models were able to produce better results than the other models investigated and could be employed successfully in modeling discharge coefficient from the available experimental data.

Evolving flexible beta basis function neural tree using extended genetic programming & Hybrid Artificial Bee Colony

In this paper, a new hybrid learning algorithm is introduced to evolve the flexible beta basis function neural tree (FBBFNT). The structure is developed using the Extended Genetic Programming (EGP) and the Beta parameters and connected weights are optimized by the Hybrid Artificial Bee Colony algorithm. This hybridization is essentially based on replacing the random Artificial Bee Colony (ABC) position with the guided Opposite-based Particle Swarm Optimization (OPSO) position. Such modification can minimize the delay which might be lead by the random position, in reaching the global solution. The performance of the proposed model is evaluated for benchmark problems drawn from time series prediction area and is compared with those of related methods.

Extension of Genetic Programming with Multiple Trees for Agent Learning

Extension of Genetic Programming with Multiple Trees for Agent Learning

Applying Dynamic Training‐Subset Selection Methods Using Genetic Programming for Forecasting Implied Volatility

Volatility is a key variable in option pricing, trading, and hedging strategies. The purpose of this article is to improve the accuracy of forecasting implied volatility using an extension of genetic programming (GP) by means of dynamic training‐subset selection methods. These methods manipulate the training data in order to improve the out‐of‐sample patterns fitting. When applied with the static subset selection method using a single training data sample, GP could generate forecasting models, which are not adapted to some out‐of‐sample fitness cases. In order to improve the predictive accuracy of generated GP patterns, dynamic subset selection methods are introduced to the GP algorithm allowing a regular change of the training sample during evolution. Four dynamic training‐subset selection methods are proposed based on random, sequential, or adaptive subset selection. The latest approach uses an adaptive subset weight measuring the sample difficulty according to the fitness cases' errors. Using real data from S&P500 index options, these techniques are compared with the static subset selection method. Based on mean squared error total and percentage of non‐fitted observations, results show that the dynamic approach improves the forecasting performance of the generated GP models, especially those obtained from the adaptive‐random training‐subset selection method applied to the whole set of training samples.

A hybrid learning algorithm for evolving Flexible Beta Basis Function Neural Tree Model

In this paper, a tree-based encoding method is introduced to represent the Beta basis function neural network. The proposed model called Flexible Beta Basis Function Neural Tree (FBBFNT) can be created and optimized based on the predefined Beta operator sets. A hybrid learning algorithm is used to evolving FBBFNT Model: the structure is developed using the Extended Genetic Programming (EGP) and the Beta parameters and connected weights are optimized by the Opposite-based Particle Swarm Optimization algorithm (OPSO). The performance of the proposed method is evaluated for benchmark problems drawn from control system and time series prediction area and is compared with those of related methods.

Use of Gene-Expression Programming to Estimate Manning’s Roughness Coefficient for High Gradient Streams

Manning’s roughness coefficient (n) has been widely used in the estimation of flood discharges or depths of flow in natural channels. Therefore, the selection of appropriate Manning’s n values is of paramount importance for hydraulic engineers and hydrologists and requires considerable experience, although extensive guidelines are available. Generally, the largest source of error in post-flood estimates (termed indirect measurements) is due to estimates of Manning’s n values, particularly when there has been minimal field verification of flow resistance. This emphasizes the need to improve methods for estimating n values. The objective of this study was to develop a soft computing model in the estimation of the Manning’s n values using 75 discharge measurements on 21 high gradient streams in Colorado, USA. The data are from high gradient (S > 0.002 m/m), cobble- and boulder-bed streams for within bank flows. This study presents Gene-Expression Programming (GEP), an extension of Genetic Programming (GP), as an improved approach to estimate Manning’s roughness coefficient for high gradient streams. This study uses field data and assessed the potential of gene-expression programming (GEP) to estimate Manning’s n values. GEP is a search technique that automatically simplifies genetic programs during an evolutionary processes (or evolves) to obtain the most robust computer program (e.g., simplify mathematical expressions, decision trees, polynomial constructs, and logical expressions). Field measurements collected by Jarrett (J Hydraulic Eng ASCE 110: 1519–1539, 1984) were used to train the GEP network and evolve programs. The developed network and evolved programs were validated by using observations that were not involved in training. GEP and ANN-RBF (artificial neural network-radial basis function) models were found to be substantially more effective (e.g., R2 for testing/validation of GEP and RBF-ANN is 0.745 and 0.65, respectively) than Jarrett’s (J Hydraulic Eng ASCE 110: 1519–1539, 1984) equation (R2 for testing/validation equals 0.58) in predicting the Manning’s n.

COMPUTATION OF DISCHARGE THROUGH SIDE SLUICE GATE USING GENE‐EXPRESSION PROGRAMMING

ABSTRACTThe existing equations for the computation of discharge through a sluice gate are derived based on the statistical approach and implicit in form. This paper presents an intelligent prediction of discharge through a side sluice gate under free flow conditions using gene‐expression programming (GEP), which is an extension of genetic programming (GP). Considering the flow through a side sluice gate as orifice flow, a closed form equation for the prediction of discharge is derived. The coefficient of discharge is related to the approximate Froude number in the main channel and the ratio of upstream depth and gate opening. The proposed equation for the coefficient of discharge using GEP and available existing data produces better results (R2 = 0.935 and RMSE = 0.0658) compared to the existing equations for the side sluice gate. The computed coefficient of discharge is within ±5% of those observed for the data used for the validation of the proposed equation. The average percentage error in the coefficient of discharge computed using proposed equation is of the order of 2.15. Copyright © 2012 John Wiley & Sons, Ltd.

Suspended sediment load prediction of river systems: GEP approach

This study presents gene expression programming (GEP), an extension of genetic programming, as an alternative approach to modeling the suspended sediment load relationship for the three Malaysian rivers. In this study, adaptive neuro-fuzzy inference system (ANFIS), regression model, and GEP approaches were developed to predict suspended load in three Malaysian rivers: Muda River, Langat River, and Kurau River [ANFIS (R 2 = 0.93, root mean square error (RMSE) = 3.19, and average error (AE) = 1.12) and regression model (R 2 = 0.63, RMSE = 13.96, and AE = 12.69)]. Additionally, the explicit formulations of the developed GEP models are presented (R 2 = 0.88, RMSE = 5.19, and AE = 6.5). The performance of the GEP model was found to be acceptable compare to ANFIS and better than the conventional models.

Gene-expression programming for transverse mixing coefficient

Summary This study presents gene-expression programming (GEP), which is an extension of genetic programming (GP), as an alternative approach to predict the transverse mixing coefficient in open channel flows. Laboratory data were collected in the present study and also from the literature for the transverse mixing coefficient covering wide range of flow conditions. These data were used for the development and testing of the proposed method. A functional relation for the estimation of transverse mixing coefficient has been developed using GEP. The proposed GEP approach produced satisfactory results compared to the existing predictors for the transverse mixing coefficient.

Gene-expression programming to predict scour at a bridge abutment

The process involved in the local scour at an abutment is so complex that it makes it difficult to establish a general empirical model to provide accurate estimation for scour. This study presents the use of gene-expression programming (GEP), which is an extension of genetic programming (GP), as an alternative approach to estimate the scour depth. The datasets of laboratory measurements were collected from the published literature and used to train the network or evolve the program. The developed network and evolved programs were validated by using the observations that were not involved in training. The proposed GEP approach gives satisfactory results compared with existing predictors and artificial neural network (ANN) modeling in predicting the scour depth at an abutment.

Gene-Expression Programming for the Development of a Stage-Discharge Curve of the Pahang River

This study presents Gene-Expression Programming (GEP), an extension of Genetic Programming (GP), as an alternative approach to modeling the stage-discharge relationship for the Pahang River. The results are compared to those obtained by more conventional methods, i.e., the stage rating curve (SRC) and regression techniques. Additionally, the explicit formulations of the developed GEP models are presented. The performance of the GEP model was found to be substantially superior to both GP and the conventional models.

Gene-Expression Programming for Sediment Transport in Sewer Pipe Systems

Old sewerage systems were designed based on empirical rules to minimize sediment problems and a list of codes for self-cleansing sewers. These codes were applicable to noncohesive sediments (typically storm sewers). This study presents gene-expression programming (GEP), which is an extension of genetic programming (GP), as an alternative approach for modeling the functional relationships of sediment transport in sewer pipe systems. A functional relation has been developed using GEP. The proposed relationship can be applied to different boundaries with partial flow. The proposed GEP approach gives satisfactory results (r2=0.97 and MSE=0.0054) compared to the existing predictor.