Version Of Genetic Programming Research Articles

Towards a Vectorial Approach to Predict Beef Farm Performance

Accurate livestock management can be achieved by means of predictive models. Critical factors affecting the welfare of intensive beef cattle husbandry systems can be difficult to be detected, and Machine Learning appears as a promising approach to investigate the hundreds of variables and temporal patterns lying in the data. In this article, we explore the use of Genetic Programming (GP) to build a predictive model for the performance of Piemontese beef cattle farms. In particular, we investigate the use of vectorial GP, a recently developed variant of GP, that is particularly suitable to manage data in a vectorial form. The experiments conducted on the data from 2014 to 2018 confirm that vectorial GP can outperform not only the standard version of GP but also a number of state-of-the-art Machine Learning methods, such as k-Nearest Neighbors, Generalized Linear Models, feed-forward Neural Networks, and long- and short-term memory Recurrent Neural Networks, both in terms of accuracy and generalizability. Moreover, the intrinsic ability of GP in performing an automatic feature selection, while generating interpretable predictive models, allows highlighting the main elements influencing the breeding performance.

Accurate High Performance Concrete Prediction with an Alignment-Based Genetic Programming System

In 2013, our research group published a contribution in which a new version of genetic programming, called Geometric Semantic Genetic Programming (GSGP), was fostered as an appropriate computational intelligence method for predicting the strength of high-performance concrete. That successful work, in which GSGP was shown to outperform the existing systems, allowed us to promote GSGP as the new state-of-the-art technology for high-performance concrete strength prediction. In this paper, we propose, for the first time, a novel genetic programming system called Nested Align Genetic Programming (NAGP). NAGP exploits semantic awareness in a completely different way compared to GSGP. The reported experimental results show that NAGP is able to significantly outperform GSGP for high-performance concrete strength prediction. More specifically, not only NAGP is able to obtain more accurate predictions than GSGP, but NAGP is also able to generate predictive models with a much smaller size, and thus easier to understand and interpret, than the ones generated by GSGP. Thanks to this ability of NAGP, we are able here to show the model evolved by NAGP, which was impossible for GSGP.

An evolutionary system for the prediction of high performance concrete strength based on semantic genetic programming

High-performance concrete, besides aggregate, cement, and water, incorporates supplementary cementitious materials, such as fly ash and blast furnace slag, and chemical admixture, such as superplasticizer. Hence, it is a highly complex material and modeling its behavior represents a difficult task. This paper presents an evolutionary system for the prediction of high performance concrete strength. The proposed framework blends a recently developed version of genetic programming with a local search method. The resulting system enables us to build a model that produces an accurate estimation of the considered parameter. Experimental results show the suitability of the proposed system for the prediction of concrete strength. The proposed method produces a lower error with respect to the state-of-the art technique. The paper provides two contributions: from the point of view of the high performance concrete strength prediction, a system able to outperform existing state-of-the-art techniques is defined; from the machine learning perspective, this case study shows that including a local searcher in the geometric semantic genetic programming system can speed up the convergence of the search process.

Dow Jones Index Return Forecasting: Semantics Based Genetic Programming with Local Search Optimizer

Making accurate stock price predictions is the pillar of effective decisions in high-velocity environments since the successful prediction of future prices could yield significant profit and reduce operational costs. Generally, solutions for this task are based on trend predictions and are driven by various factors. To add to the existing body of knowledge, we propose a semantics-based genetic programming framework. The proposed framework blends a recently developed version of genetic programming that uses semantic genetic operators with a local search method. To analyse the appropriateness of the proposed computational method for stock market price prediction, we analysed data related to the Dow Jones index and to the Istanbul Stock Index. Experimental results confirm the suitability of the proposed method for predicting stock market prices. In fact, the system produces lower errors with respect to the existing state-of-the art techniques, such as neural networks and support vector machines. forecasting; financial markets; genetic programming; semantics; local search.

An evolutionary system for exploitation of fractured geothermal reservoirs

In this work, we present a novel methodology to integrate one of the most advanced technique for modeling fractured media for underground systems with a semantics-based genetic programming technique. The objective of the study is to develop a global framework to forecast the temperature of fractured reservoirs. The numerical method used to solve the physical equations is able to handle different fracture distributions without changing the background computational grid, i.e. the mesh of the rock matrix, as well as letting geometrically uncoupled the one co-dimensional fracture meshes. In the context of temperature forecasting, the use of a recently defined variant of genetic programming is taken into account for finding (quasi-)perfect solutions with high probability and for generating models able to produce near optimal predictions also on unseen data. The proposed computational intelligence technique integrates, in a recently developed version of genetic programming that uses semantic genetic operators, a “greedy” crossover and a self tuning algorithm. Experimental results confirm the suitability of the proposed method in predicting the correct temperature distribution in probes inside the domain.

Faster GPU-based genetic programming using a two-dimensional stack

Genetic programming (GP) is a computationally intensive technique which also has a high degree of natural parallelism. Parallel computing architectures have become commonplace especially with regards to Graphics Processing Units (GPU). Hence, versions of GP have been implemented that utilise these highly parallel computing platforms enabling significant gains in the computational speed of GP to be achieved. However, recently a two-dimensional stack approach to GP using a multi-core CPU also demonstrated considerable performance gains. Indeed, performances equivalent to or exceeding that achieved by a GPU were demonstrated. This paper will demonstrate that a similar two-dimensional stack approach can also be applied to a GPU-based approach to GP to better exploit the underlying technology. Performance gains are achieved over a standard single-dimensional stack approach when utilising a GPU. Overall, a peak computational speed of over 55 billion Genetic Programming Operations per Second are observed, a twofold improvement over the best GPU-based single-dimensional stack approach from the literature.

Improving genetic programming for the prediction of pharmacokinetic parameters

The prediction of pharmacokinetic parameters is a crucial phase of the drug discovery process, and the automatization of this task is a hot topic in computational bio-medicine. In the last 10 years, a significant amount of research has been published reporting on applications of genetic programming to the prediction of pharmacokinetic parameters. This paper summarizes and discusses some of those contributions. In particular, the focus is on the idea that pharmacokinetic problems are so complex that the “canonic” version of genetic programming is often not able to perform appropriately on them. At the same time, genetic programming has a high degree of versatility, given by the opportunity it offers of adapting many crucial parts of its algorithm, among which the fitness function and the employed genetic operators. This gives us the chance to improve standard genetic programming in several different ways. For instance, sophisticated fitness functions, methods to control bloat and operators to exploit the geometry of the semantic space are discussed here.

Practical performance models of algorithms in evolutionary program induction and other domains

Evolutionary computation techniques have seen a considerable popularity as problem solving and optimisation tools in recent years. Theoreticians have developed a variety of both exact and approximate models for evolutionary program induction algorithms. However, these models are often criticised for being only applicable to simplistic problems or algorithms with unrealistic parameters. In this paper, we start rectifying this situation in relation to what matters the most to practitioners and users of program induction systems: performance. That is, we introduce a simple and practical model for the performance of program-induction algorithms. To test our approach, we consider two important classes of problems — symbolic regression and Boolean function induction — and we model different versions of genetic programming, gene expression programming and stochastic iterated hill climbing in program space. We illustrate the generality of our technique by also accurately modelling the performance of a training algorithm for artificial neural networks and two heuristics for the off-line bin packing problem.We show that our models, besides performing accurate predictions, can help in the analysis and comparison of different algorithms and/or algorithms with different parameters setting. We illustrate this via the automatic construction of a taxonomy for the stochastic program-induction algorithms considered in this study. The taxonomy reveals important features of these algorithms from the performance point of view, which are not detected by ordinary experimentation.

USING DYNAMIC FORECASTING GENETIC PROGRAMMING (DFGP) TO FORECAST UNITED STATES GROSS DOMESTIC PRODUCT (US GDP) WITH MILITARY EXPENDITURE AS AN EXPLANATORY VARIABLE

Classic time‐series forecasting models can be divided into exponential smoothing, regression, ARIMA, threshold, and GARCH models. Functional form is investigator‐specified, and all methods assume that the data generation process across all segments of the examined time‐series is constant. In contrast, the aim of heuristic methods is to automate the discovery of functional form and permit different segments of a time‐series to stem from different underlying data generation processes. These methods are categorized into those based on neural networks (NN) and those based on evolutionary computation, the latter further divided into genetic algorithms (GA), evolutionary programming (EP), and genetic programming (GP). However, the duration of the time‐series itself is still investigator determined. This paper uses a dynamic forecasting version of GP (DFGP), where even the length of the time‐series is automatically discovered. The method is applied to an examination of US GDP that includes military expenditure among its determinants and is compared to a regression‐based forecast. We find that DFGP and a regression‐based forecast yield comparable results but with the significant proviso that DFGP does not make any prior assumption about functional form or the time‐span from which forecasts are produced.