Implementation Of Genetic Programming Research Articles

Faulty diagnostics model in e-commerce using AI

Risk administration has increased due to increased online orders to avoid sales invoices. Failure to pay a bill within 90 days of receipt constitutes overdue payment. The credit rating is used to determine the probability that consumers will fail. The CR has been thoroughly studied and several learning algorithms have been proposed. The main goal is to create a CR model for the role of Risk Solution Services (RSS), an industry standard for predicting the parameters of customer default in e-commerce risk assessment. The risk assessment included a general concept, exclusion criteria, and details of the ordering process. The most recent design should operate both independently and in conjunction with the NRC Primary Risk Audit as it is intended to replace the overall screening risk assessment. This article is about a CR implementation of Genetic Programming (GP) with Artificial Intelligence (AI). The dataset includes RSS-enabled purchase requisitions. The results show that the GP pre-risk control model goes beyond the generic CR model in terms of classification accuracy. A system with more discriminatory capacity is produced by combining GP models with the NRC's major risk assessment.

Genetic programming-based regression for temporal data

Various machine learning techniques exist to perform regression on temporal data with concept drift occurring. However, there are numerous nonstationary environments where these techniques may fail to either track or detect the changes. This study develops a genetic programming-based predictive model for temporal data with a numerical target that tracks changes in a dataset due to concept drift. When an environmental change is evident, the proposed algorithm reacts to the change by clustering the data and then inducing nonlinear models that describe generated clusters. Nonlinear models become terminal nodes of genetic programming model trees. Experiments were carried out using seven nonstationary datasets and the obtained results suggest that the proposed model yields high adaptation rates and accuracy to several types of concept drifts. Future work will consider strengthening the adaptation to concept drift and the fast implementation of genetic programming on GPUs to provide fast learning for high-speed temporal data.

Estimation of COVID-19 epidemic curves using genetic programming algorithm.

This paper investigates the possibility of the implementation of Genetic Programming (GP) algorithm on a publicly available COVID-19 data set, in order to obtain mathematical models which could be used for estimation of confirmed, deceased, and recovered cases and the estimation of epidemiology curve for specific countries, with a high number of cases, such as China, Italy, Spain, and USA and as well as on the global scale. The conducted investigation shows that the best mathematical models produced for estimating confirmed and deceased cases achieved R2 scores of 0.999, while the models developed for estimation of recovered cases achieved the R2 score of 0.998. The equations generated for confirmed, deceased, and recovered cases were combined in order to estimate the epidemiology curve of specific countries and on the global scale. The estimated epidemiology curve for each country obtained from these equations is almost identical to the real data contained within the data set.

A practical design of hash functions for IPv6 using multi-objective genetic programming

Hash functions are widely used in high-speed network traffic measurement. A hash function of high quality is supposed to meet the requirements of collision free and fast execution. Existing works have already developed methods to generate hash functions for IPv4 data, while IPv6 data with much longer addresses and different data characteristics may decline the effectiveness of those methods. In this paper, we present a practical design of hash functions for IPv6 measurement, based on the entropy analysis of IPv6 network data and an automated method of multi-objective genetic programming (GP). Considering our specific application of hash functions, we use three fitness functions as the optimization objectives, including active flow estimation, uniformity and seed avalanche effect, among which the active flow estimation is the main objective as the specific measurement task. In implementation of multi-objective GP, we adopted a strategy to limit the hash functions to shorter execution time than other hash functions by advanced experimental investigation. Experiments were conducted to construct hash functions for WIDE IPv6 network data. The results show that our generated hash functions have high usability on different evaluation criteria. It indicates that our generated hash functions are superior in active flow estimation and execution time and could compete with state of art hash functions in terms of uniformity and generating independent hash values for data structures like Bloom Filter.

Evolutionary music: applying evolutionary computation to the art of creating music

We present a review of the application of genetic programming (GP) and other variations of evolutionary computation (EC) to the creative art of music composition. Throughout the development of EC methods, since the early 1990s, a small number of researchers have considered aesthetic problems such as the act of composing music alongside other more traditional problem domains. Over the years, interest in these aesthetic or artistic domains has grown significantly. We review the implementation of GP and EC for music composition in terms of the compositional task undertaken, the algorithm used, the representation of the individuals and the fitness measure employed. In these aesthetic studies we note that there are more variations or generalisations in the algorithmic implementation in comparison to traditional GP experiments; even if GP is not explicitly stated, many studies use representations that are distinctly GP-like. We determine that there is no single compositional challenge and no single best evolutionary method with which to approach the act of music composition. We consider autonomous composition as a computationally creative act and investigate the suitability of EC methods to the search for creativity. We conclude that the exploratory nature of evolutionary methods are highly appropriate for a wide variety of compositional tasks and propose that the development and study of GP and EC methods on creative tasks such as music composition should be encouraged.

Inference of compact nonlinear dynamic models by epigenetic local search

We introduce a method to enhance the inference of meaningful dynamic models from observational data by genetic programming (GP). This method incorporates an inheritable epigenetic layer that specifies active and inactive genes for a more effective local search of the model structure space. We define several GP implementations using different features of epigenetics, such as passive structure, phenotypic plasticity, and inheritable gene regulation. To test these implementations, we use hundreds of data sets generated from nonlinear ordinary differential equations (ODEs) in several fields of engineering and from randomly constructed nonlinear ODE models. The results indicate that epigenetic hill climbing consistently produces more compact dynamic equations with better fitness values, and that it identifies the exact solution of the system more often, validating the categorical improvement of GP by epigenetic local search. The results further indicate that when faced with complex dynamics, epigenetic hill climbing reduces the computational effort required to infer the correct underlying dynamics. We then apply the method to the identification of three real-world systems: a cascaded tanks system, a chemical distillation tower, and an industrial wind turbine. We analyze its solutions in comparison to theoretical and black-box approaches in terms of accuracy and intelligibility. Finally, we analyze population homology to evaluate the efficiency of the method. The results indicate that the epigenetic implementations provide protection from premature convergence by maintaining diversity in silenced portions of programs.

An Efficient New Scheme of Fitness Evaluation in Genetic Programming using the R Language

The aim of this paper is to propose and analyze several fitness evaluation schemes for solving regression problems using Genetic Programming. The proposed schemes are designed considering the particularities and characteristics of the R Language, and particularly the capacities of the language for matrix manipulation and mathematical expressions evaluation. Experimental results show that some the proposed schemes are able to reduce until 99% the time spend in fitness evaluation in comparison with the original genetic programming implementation using a traditional tree structure.

Improving the performance of GPU-based genetic programming through exploitation of on-chip memory

Genetic Programming (GP) (Koza, Genetic programming, MIT Press, Cambridge, 1992) is well-known as a computationally intensive technique. Subsequently, faster parallel versions have been implemented that harness the highly parallel hardware provided by graphics cards enabling significant gains in the performance of GP to be achieved. However, extracting the maximum performance from a graphics card for the purposes of GP is difficult. A key reason for this is that in addition to the processor resources, the fast on-chip memory of graphics cards needs to be fully exploited. Techniques will be presented that will improve the performance of a graphics card implementation of tree-based GP by better exploiting this faster memory. It will be demonstrated that both L1 cache and shared memory need to be considered for extracting the maximum performance. Better GP program representation and use of the register file is also explored to further boost performance. Using an NVidia Kepler 670GTX GPU, a maximum performance of 36 billion Genetic Programming Operations per Second is demonstrated.

Scalable Evolutionary Computation for Efficient Information Extraction from Remote Sensed Imagery

Abstract. Evolutionary computation, in the form of genetic programming, is used to aid information extraction process from high-resolution satellite imagery in a semi-automatic fashion. Distributing and parallelizing the task of evaluating all candidate solutions during the evolutionary process could significantly reduce the inherent computational cost of evolving solutions that are composed of multichannel large images. In this study, we present the design and implementation of a system that leverages cloud-computing technology to expedite supervised solution development in a centralized evolutionary framework. The system uses the MapReduce programming model to implement a distributed version of the existing framework in a cloud-computing platform. The proposed system has two major subsystems; (i) data preparation: the generation of random spectral indices; and (ii) distributed processing: the distributed implementation of genetic programming, which is used to spectrally distinguish the features of interest from the remaining image background in the cloud computing environment in order to improve scalability. The proposed system reduces response time by leveraging the vast computational and storage resources in a cloud computing environment. The results demonstrate that distributing the candidate solutions reduces the execution time by 91.58%. These findings indicate that such technology could be applied to more complex problems that involve a larger population size and number of generations.

Balancing exploration and exploitation in genetic programming using inversion with individualized self-adaptation

In this article we explore and develop a holistic scheme of self adaptive, individualized genetic operators combined with an adaptive tournament size together with a novel implementation of an inversion genetic operator which is suitable for tree based Genetic Programming. We test this scheme on several benchmark Binary Classification problems and find that the proposed techniques deliver superior performance when compared with both a tuned GP configuration and a feedback adaptive Genetic Programming implementation. Our results also demonstrate that an inversion operator may have a useful role to play in exploitation, particularly when used towards the end of evolution to facilitate gradual convergence of the learning system towards a good solution.

Incorporating basic hydrological concepts into genetic programming for rainfall-runoff forecasting

This paper focuses on improving rainfall-runoff forecasts by a combination of genetic programming (GP) and basic hydrological modelling concepts. GP is a general optimisation technique for making an automated search of a computer program that solves some particular problem. The SORD! program was developed for the purposes of this study (in the R programming language). It is an implementation of canonical GP. Special functions are used for a combined approach of hydrological concepts and GP. The special functions are a reservoir model, a simple moving average model, and a cumulative sum and delay operator. The efficiency of the approach presented here is tested on runoff predictions for five catchments of various sizes. The input data consists of daily rainfall and runoff series. The forecast step is one day. The performance of the proposed approach is compared with the results of the artificial neural network model (ANN) and with the GP model without special functions. GP combined with these concepts provides satisfactory performance, and the simulations seem to be more accurate than the results of ANN and GP without these functions. An additional advantage of the proposed approach is that it is not necessary to determine the input lag, and there is better convergence. The SORD! program provides an easy-to-use alternative for data-oriented modelling combined with simple concepts used in hydrological modelling.

A Genetic Programming Approach for EUR/USD Exchange Rate Forecasting and Trading

The purpose of this article is to present a novel genetic programming trading technique in the task of forecasting the next day returns when trading the EUR/USD exchange rate based on the exchange rates of historical data. Aiming at testing its effectiveness, we benchmark the forecasting performance of our genetic programming implementation with three traditional strategies (naive strategy, MACD, and a buy & hold strategy) plus a hybrid evolutionary artificial neural network approach. The proposed genetic programming technique was found to demonstrate the highest trading performance in terms of annualized return and information ratio when compared to all other strategies which have been used. When more elaborate trading techniques, such as leverage, were combined with the examined models, the genetic programming approach still presented the highest trading performance. To the best of our knowledge, this is the first time that genetic programming is applied in the problem of effectively modeling and trading with the EUR/USD exchange rate. Our application now offers practitioners with an effective and extremely promising set of results when forecasting in the foreign exchange market. The developed genetic programming environment is implemented using the C++ programming language and includes a variation of the genetic programming algorithm with tournament selection.

Fast parallel genetic programming: multi-core CPU versus many-core GPU

Genetic Programming (GP) is a computationally intensive technique which is also highly parallel in nature. In recent years, significant performance improvements have been achieved over a standard GP CPU-based approach by harnessing the parallel computational power of many-core graphics cards which have hundreds of processing cores. This enables both fitness cases and candidate solutions to be evaluated in parallel. However, this paper will demonstrate that by fully exploiting a multi-core CPU, similar performance gains can also be achieved. This paper will present a new GP model which demonstrates greater efficiency whilst also exploiting the cache memory. Furthermore, the model presented in this paper will utilise Streaming SIMD Extensions to gain further performance improvements. A parallel version of the GP model is also presented which optimises multiple thread execution and cache memory. The results presented will demonstrate that a multi-core CPU implementation of GP can yield performance levels that match and exceed those of the latest graphics card implementations of GP. Indeed, a performance gain of up to 420-fold over standard GP is demonstrated and a threefold gain over a graphics card implementation.

Asynchronous migration for parallel genetic programming on a computer cluster with multi-core processors

An island model is a typical implementation of genetic programming on parallel computers with distributed memory. The island model has a migration facility that sends/receives some individuals in an island to/from another island to maintain diversity. The island model requires synchronization to migrate same-generation individuals between islands, and this synchronization causes an increase in computation time. This article proposes a new parallel genetic programming implementation based on the island model with asynchronous migration. Most recent computers are equipped with one or more multi-core processors, and are suitable for multi-threading. Therefore we employ a communication thread for migration between islands. The communication thread on a processor communicates with the communication thread on another processor to migrate individuals at appropriate intervals. Since the migration and other genetic operations can be independently processed on each core, and since we allow the exchange of individuals of different generations, no synchronization is needed in our implementation. In addition, a fitness calculation is also executed in parallel by the remaining cores. Experimental results show that the proposed method can reduce the computation time to about 17% in serial GP by using 40 threads.

Stock trading strategy creation using GP on GPU

This paper investigates the speed improvements available when using a graphics processing unit (GPU) for evaluation of individuals in a genetic programming (GP) environment. An existing GP system is modified to enable parallel evaluation of individuals on a GPU device. Several issues related to implementing GP on GPU are discussed, including how to perform tree-based GP on a device without recursion support, as well as the effect that proper memory layout can have on speed increases when using CUDA-enabled nVidia GPU devices. The specific GP implementation is designed to evolve stock trading strategies using technical analysis indicators. The second goal of this research is to investigate the possible improvement in performance when training individuals on a larger number of stocks and training days. This increased training size (nearly 100,000 training points) is enabled due to the speedups realized by GPU evaluation. Several different scenarios were used to test various speed optimizations of GP evaluation on the GPU device, with a peak speedup factor of over 600 (when compared to sequential evaluation on a 2.4 GHz CPU). Also, it is found that increasing the number of stocks and the length of the training period can result in higher out-of-training testing profitability.

An elaborate education of basic genetic programming using C++

Evolutionary search is a global search method based on natural selection. In engineering curriculum, these techniques are taught in courses like Evolutionary Computation, Engineering Optimization, etc. Genetic algorithm (GA) is popular among these algorithms. Genetic programming (GP), developed by John Koza, is a powerful extension of GA where a chromosome/computer program (CP) is coded as a rooted point-labeled tree with ordered branches. The search space is the space of all possible CPs (trees) consisting of functions and terminals appropriate to the problem domain. GP uses, like GA, crossover and mutation for evolution. Due to tree-structured coding of individuals, the initial population generation, genetic operators' use, and tree decoding for fitness evaluations demand careful computer programming. This article describes the programming steps of GP implementation (using C++ language) for students' easy understanding with pseudocodes for each step. Two application examples are also illustrated. © 2009 Wiley Periodicals, Inc. Comput Appl Eng Educ 18: 434–448, 2010; View this article online at wileyonlinelibrary.com; DOI 10.1002/cae.20165

Genetic Programming for the Automatic Design of Controllers for a Surface Ship

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this paper, the implementation of genetic programming (GP) to design a controller structure is assessed. GP is used to evolve control strategies that, given the current and desired state of the propulsion and heading dynamics of a supply ship as inputs, generate the commanded forces required to maneuver the ship. The controllers created using GP are evaluated through computer simulations and real maneuverability tests in a laboratory water basin facility. The robustness of each controller is analyzed through the simulation of environmental disturbances. In addition, GP runs in the presence of disturbances are carried out so that the different controllers obtained can be compared. The particular vessel used in this paper is a scale model of a supply ship called CyberShip II. The results obtained illustrate the benefits of using GP for the automatic design of propulsion and navigation controllers for surface ships. </para>

Thalassaemia classification by neural networks and genetic programming

This paper presents the use of a neural network and a decision tree, which is evolved by genetic programming (GP), in thalassaemia classification. The aim is to differentiate between thalassaemic patients, persons with thalassaemia trait and normal subjects by inspecting characteristics of red blood cells, reticulocytes and platelets. A structured representation on genetic algorithms for non-linear function fitting or STROGANOFF is the chosen architecture for genetic programming implementation. For comparison, multilayer perceptrons are explored in classification via a neural network. The classification results indicate that the performance of the GP-based decision tree is approximately equal to that of the multilayer perceptron with one hidden layer. But the multilayer perceptron with two hidden layers, which is proven to have the most suitable architecture among networks with different number of hidden layers, outperforms the GP-based decision tree. Nonetheless, the structure of the decision tree reveals that some input features have no effects on the classification performance. The results confirm that the classification accuracy of the multilayer perceptron with two hidden layers can still be maintained after the removal of the redundant input features. Detailed analysis of the classification errors of the multilayer perceptron with two hidden layers, in which a reduced feature set is used as the network input, is also included. The analysis reveals that the classification ambiguity and misclassification among persons with minor thalassaemia trait and normal subjects is the main cause of classification errors. These results suggest that a combination of a multilayer perceptron with a blood cell analysis may give rise to a guideline/hint for further investigation of thalassaemia classification.

A Jxta Based Asynchronous Peer-to-Peer Implementation of Genetic Programming

Solving complex real-world problems using evolutionary computation is a CPU time-consuming task that requires a large amount of computational resources. Peerto-Peer (P2P) computing has recently revealed as a powerful way to harness these resources and efficiently deal with such problems. In this paper, we present P-CAGE: a P2P environment for Genetic Programming based on the JXTA protocols. P-CAGE is based on a hybrid multi-island model that combines the island model with the cellular model. Each island adopts a cellular model and the migration occurs between neighboring peers placed in a virtual ring topology. Three different termination criteria (effort, time and maxgen) have been implemented. Experiments were conducted on some popular benchmarks and scalability, accuracy and the effect of migration have been studied. Performance are at least comparable with classical distributed models, retaining the obvious advantages in terms of decentralization, fault tolerance and scalability of P2P systems. We also demonstrated the important effect of migration in accelerating the convergence.

Nonlinear ranking function representations in genetic programming-based ranking discovery for personalized search

Ranking function is instrumental in affecting the performance of a search engine. Designing and optimizing a search engine's ranking function remains a daunting task for computer and information scientists. Recently, genetic programming (GP), a machine learning technique based on evolutionary theory, has shown promise in tackling this very difficult problem. Ranking functions discovered by GP have been found to be significantly better than many of the other existing ranking functions. However, current GP implementations for ranking function discovery are all designed utilizing the Vector Space model in which the same term weighting strategy is applied to all terms in a document. This may not be an ideal representation scheme at the individual query level considering the fact that many query terms should play different roles in the final ranking. In this paper, we propose a novel nonlinear ranking function representation scheme and compare this new design to the well-known Vector Space model. We theoretically show that the new representation scheme subsumes the traditional Vector Space model representation scheme as a special case and hence allows for additional flexibility in term weighting. We test the new representation scheme with the GP-based discovery framework in a personalized search (information routing) context using a TREC web corpus. The experimental results show that the new ranking function representation design outperforms the traditional Vector Space model for GP-based ranking function discovery.