Cartesian Genetic Programming Research Articles

GA evolved CGP configuration data for digital circuit design on embryonic architecture

Embryonic architecture that carries self-evolving design with fault tolerant feature is proposed for deep space missions. Fault tolerance is achieved in the embryonic architecture due to its homogeneous structure. The cloning of configuration data or genome data to all the embryonic cells makes each cell capable of selecting required cell function using selective gene. The primary digital circuits of avionics are implemented on the fabric, where the configuration data in Cartesian Genetic Programming (CGP) format is evolved through customized GA. The CGP format is preferred over LUT format for the circuit configuration data due to its fixed data size in case of modular design. Further the CGP format enables fault detection at embryonic cell level as well as logic gate level. The various combinational and sequential circuits like adder, comparator, multiplier, register and counter are designed and implemented on embryonic fabric using Verilog. The circuit performance is evaluated using simulation. The proposed PHsClone genetic algorithm (GA) design with parallel-pipeline approach is to achieve faster convergence. Four concurrent PHsClone GA executions (four parallel threads) achieve convergence for the 10 times faster for a 1-bit adder, and 3 times faster for a 2-bit comparator.

Continuous Cartesian Genetic Programming based representation for multi-objective neural architecture search

We propose a novel neural architecture search (NAS) approach for the challenge of designing convolutional neural networks (CNNs) that achieve a good tradeoff between complexity and accuracy. We rely on Cartesian genetic programming (CGP) and integrated real-based and block-chained CNN representation, for optimization using multi-objective evolutionary algorithms (MOEAs) in the continuous domain. We introduce two variants, CGP-NASV1 and CGP-NASV2, which differ in the granularity of their respective search spaces. To evaluate the proposed algorithms, we utilized the non-dominated sorting genetic algorithm II (NSGA-II) on the CIFAR-10, CIFAR-100,and SVHN datasets. Additionally, we extended the empirical analysis while maintaining the same solution representation to assess other searching techniques such as differential evolution (DE), the multi-objective evolutionary algorithm based on decomposition (MOEA/D), and the S metric selection evolutionary multi-objective algorithm (SMS-EMOA). The experimental results demonstrate that our approach exhibits competitive classification performance and model complexity compared to state-of-the-art methods.

An adaptive mutation for cartesian genetic programming using an $$\epsilon $$-greedy strategy

An adaptive mutation for cartesian genetic programming using an $$\epsilon $$-greedy strategy

W-AMA: Weight-aware Approximate Multiplication Architecture for neural processing

This paper presents the Weight-aware Approximate Multiplication Architecture (W-AMA) for Deep Neural Networks (DNNs). Considering the Gaussian-like weight distribution, it deploys an accuracy-configurable computing component to improve the computational efficiency. Two techniques for effectively integrating the W-AMA into DNN accelerator are presented: (1) A Cartesian Genetic Programming (CGP) based approximate multiplier is designed and selectable to compute the Least Significant Bit (LSB) for a higher accuracy mode. The Reward–Penalty-Coefficient (RPC) is proposed to achieve the internal-compensation. (2) The Hessian-Aware-Approximation (HAA) method is utilized for hybrid approximate modes cross-layer mapping. Based on the W-AMA, an energy-efficient DNN accelerator is proposed and evaluated on 28 nm technology. It can achieve the energy efficiency of 9.6 TOPS/W, and the computational energy efficiency can be improved by 1.5× compared with the standard units, with an 0.52% accuracy loss on CIFAR-10 using ResNet-18.

Classification of normal and abnormal heart by classifying PCG signal using MFCC coefficients and CGP-ANN classifier

Globally, A leading cause of death is heart disease and a serious public health concern. The anomalies in heart sound appears before the heart disease symptoms. The sounds are type of auscultation, which is a process dealing with sounds in a body that generates due to mechanical vibrations of organs, Auscultation is a potential method in medical science to detect abnormalities in heart sounds and in case of suspicion The patient follows up with a referral for other evaluations, such as an electrocardiogram. In medical sciences early detection of symptoms are of major importance, this research work is a good step toward the detection of abnormalities in heart before symptom appearing by processing the phonocardiogram (PCG) signal. In this paper PCG signals are classified by utilizing the features of Mel frequency cepstral coefficients (MFCC) through Cartesian Genetic Programming - Artificial Network (CGP-ANN) Classifier. The diagnostic accuracy of proposed methodology is found 99.50% which is more than other classifiers like Support Vector Machine (SVM) and Convolutional Neural Network (CNN). The accuracy of model as compared to other models can prove the performance superiority of the proposed system.

Advanced Genetic Programming vs. State-of-the-Art AutoML in Imbalanced Binary Classification

The objective of this article is to provide a comparative analysis of two novel genetic programming (GP) techniques, differentiable Cartesian genetic programming for artificial neural networks (DCGPANN) and geometric semantic genetic programming (GSGP), with state-of-the-art automated machine learning (AutoML) tools, namely Auto-Keras, Auto-PyTorch and Auto-Sklearn. While all these techniques are compared to several baseline algorithms upon their introduction, research still lacks direct comparisons between them, especially of the GP approaches with state-of-the-art AutoML. This study intends to fill this gap in order to analyze the true potential of GP for AutoML. The performances of the different tools are assessed by applying them to 20 benchmark datasets of the imbalanced binary classification field, thus an area that is a frequent and challenging problem. The tools are compared across the four categories average performance, maximum performance, standard deviation within performance, and generalization ability, whereby the metrics F1-score, G-mean, and AUC are used for evaluation. The analysis finds that the GP techniques, while unable to completely outperform state-of-the-art AutoML, are indeed already a very competitive alternative. Therefore, these advanced GP tools prove that they are able to provide a new and promising approach for practitioners developing machine learning (ML) models. Doi: 10.28991/ESJ-2023-07-04-021 Full Text: PDF

Cartesian genetic programming for image analysis of the developing drosophila eye

Automatic feature extraction methods have gained increasing attention in modern image processing. The confocal images of the single-layered epithelium of the developing Drosophila eye may form an excellent model system to develop methods for complex feature extraction. The aim of this work was to explore Cartesian genetic programming for determination of the boundaries of ommatidia, the light-sensitive units in the presumptive eye region. Application of Cartesian genetic programming for the analysis of Fasciclin III expression has shown good results. This opens interesting perspectives for further use of this technology in the automatic analysis of confocal images.

Cartesian Genetic Programming for Image Analysis of the Developing Drosophila Eye

Cartesian Genetic Programming for Image Analysis of the Developing Drosophila Eye

Enhancing Local Decisions in Agent-Based Cartesian Genetic Programming by CMA-ES

Cartesian genetic programming is a popular version of classical genetic programming, and it has now demonstrated a very good performance in solving various use cases. Originally, programs evolved by using a centralized optimization approach. Recently, an algorithmic level decomposition of program evolution has been introduced that can be solved by a multi-agent system in a fully distributed manner. A heuristic for distributed combinatorial problem-solving was adapted to evolve these programs. The applicability of the approach and the effectiveness of the used multi-agent protocol as well as of the evolved genetic programs for the case of full enumeration in local agent decisions has already been successfully demonstrated. Symbolic regression, n-parity, and classification problems were used for this purpose. As is typical of decentralized systems, agents have to solve local sub-problems for decision-making and for determining the best local contribution to solving program evolution. So far, only a full enumeration of the solution candidates has been used, which is not sufficient for larger problem sizes. We extend this approach by using CMA-ES as an algorithm for local decisions. The superior performance of CMA-ES is demonstrated using Koza’s computational effort statistic when compared with the original approach. In addition, the distributed modality of the local optimization is scrutinized by a fitness landscape analysis.

진화연산을 사용한 널뛰기지수 기반 강수예측 모델링

This paper investigates the jumpiness index based modeling for precipitation prediction. The jumpiness index (JI) represents different forecast jumps and the inconsistency correlation for forecasts issued at different times but valid for the same time. Unlike researches that improve the accuracy of predictions mainly, this study attempts to analyze the characteristics of prediction consistency for the numerical model GDAPS using the jumpiness index. In this work, we try to minimize prediction errors through a combination of JI elements and symbolic functions using Cartesian Genetic Programming (CGP), which is a kind of evolutionary computation to solve nonlinear regression problems. The relationship between the time-by-time jumpiness indices and the prediction error is analyzed. Also, the applicability of a combination of jumpiness indices to improve the accuracy of future forecasts is examined. Experiments are conducted to model the relationship between JI and UM error values calculated by GDAPS data for 2013 and 2014 for ground precipitation data using CGP-based symbolic nonlinear regression.

A Comparative Study of Ensemble Techniques Based on Genetic Programming: A Case Study in Semantic Similarity Assessment

The challenge of assessing semantic similarity between pieces of text through computers has attracted considerable attention from industry and academia. New advances in neural computation have developed very sophisticated concepts, establishing a new state of the art in this respect. In this paper, we go one step further by proposing new techniques built on the existing methods. To do so, we bring to the table the stacking concept that has given such good results and propose a new architecture for ensemble learning based on genetic programming. As there are several possible variants, we compare them all and try to establish which one is the most appropriate to achieve successful results in this context. Analysis of the experiments indicates that Cartesian Genetic Programming seems to give better average results.

Using Cartesian Genetic Programming Approach with New Crossover Technique to Design Convolutional Neural Networks

Using Cartesian Genetic Programming Approach with New Crossover Technique to Design Convolutional Neural Networks

Evolutionary convolutional neural network for image classification based on multi-objective genetic programming with leader–follower mechanism

As a popular research in the field of artificial intelligence in the last 2 years, evolutionary neural architecture search (ENAS) compensates the disadvantage that the construction of convolutional neural network (CNN) relies heavily on the prior knowledge of designers. Since its inception, a great deal of researches have been devoted to improving its associated theories, giving rise to many related algorithms with pretty good results. Considering that there are still some limitations in the existing algorithms, such as the fixed depth or width of the network, the pursuit of accuracy at the expense of computational resources, and the tendency to fall into local optimization. In this article, a multi-objective genetic programming algorithm with a leader–follower evolution mechanism (LF-MOGP) is proposed, where a flexible encoding strategy with variable length and width based on Cartesian genetic programming is designed to represent the topology of CNNs. Furthermore, the leader–follower evolution mechanism is proposed to guide the evolution of the algorithm, with the external archive set composed of non-dominated solutions acting as the leader and an elite population updated followed by the external archive acting as the follower. Which increases the speed of population convergence, guarantees the diversity of individuals, and greatly reduces the computational resources. The proposed LF-MOGP algorithm is evaluated on eight widely used image classification tasks and a real industrial task. Experimental results show that the proposed LF-MOGP is comparative with or even superior to 35 existing algorithms (including some state-of-the-art algorithms) in terms of classification error and number of parameters.

Speaker Verification System Based On Cartesian Genetic Programming (CGP) And Cartesian Genetic Programming Evolved ANN (CGP-E-ANN)

Speaker Verification System Based On Cartesian Genetic Programming (CGP) And Cartesian Genetic Programming Evolved ANN (CGP-E-ANN)

Machine-Made Synthesis of Stabilization System by Modified Cartesian Genetic Programming

A numerical solution of the problem of the general synthesis of a stabilization system by a symbolic regression method is considered. The goal is to automatically find a feedback control function using a computer so that the control object can reach a given terminal position from anywhere in a given region of the initial conditions with an optimal value of the quality criterion. Usually, the control synthesis problem is solved analytically or technically taking into account the specific properties of the mathematical model. We suppose that modern numerical approaches of symbolic regression can be applied to find a solution without reference to specific model equations. It is proposed to use the numerical method of Cartesian genetic programming (CGP). It was developed for automatic writing of programs but has never been used to solve the synthesis problem. In the present work, the method was modified with the principle of small variations in order to reduce the search area and increase the rate of convergence. To apply the general principle of small variations to CGP, we developed special types of variations and coding. The modified CGP searches for the mathematical expression of the feedback control function in the form of a code and, at the same time, the optimal value of the parametric vector which is also a new feature-simultaneous tuning of the parameters inside the search process. This approach enables working with objects and functions of any type, which is not always possible with analytical methods. The need to use the received solution on the onboard processor of the control object imposes certain restrictions on the used basic set of elementary functions. This article proposes the theoretical foundations of the study of these functions, and the concept of the space of machine-made functions is introduced. The capabilities of the approach are demonstrated on the numerical solution of the control system synthesis problems for a mobile robot and a Duffing model.

Julian Francis Miller, 1955-2022.

Julian Francis Miller1It is with great sadness that we report the death of our colleague and friend, Julian Miller.Julian’s work is well known throughout the Artificial Life community: His Cartesian genetic programming (CGP) and in materio computing are foundational concepts. He also made contributions in morphological computing and neurocomputing, all based on his fascination with evolution as a means of attacking and solving problems. Like many in the ALife community, he had an interdisciplinary career, commencing with a first degree in Physics and a PhD in Mathematics, followed by research in Natural Computing and material computing at the universities of Napier, Birmingham, and York in the UK.Julian invented CGP (Miller, 1999), a way of encoding graph programs (functional nodes connected by edges) in a string of integers, allowing the string to be evolved in the standard way, with the graph (located on a Cartesian grid, hence its name) produced as the result of a genotype to phenotype mapping. From this simple beginning, Julian and his students continued to develop the approach, and other researchers joined in. Ten years later, the field had grown significantly, with many researchers both using CGP in their own work and extending the original concept. Indeed, the field had grown enough that Julian could edit an entire book on the topic (Miller, 2011). Ten years later still, the field shows no signs of abating, and Julian wrote a 40-page review for Genetic Programming and Evolvable Machines on CGP’s status, its many variants, and its future prospects (Miller, 2020).Julian was also a pioneer in the field of in materio computing (Miller & Downing, 2002), which exploits the physical properties of unconventional materials, such as liquid crystals (Harding & Miller, 2004) and carbon nanotubes (Miller et al., 2014), to perform computation intrinsically, in what he dubbed a “Field Programmable Matter Array.” His original work used evolutionary algorithms directly to configure the materials. Later, he also used Reservoir Computing as a more abstract model for getting these materials to compute (Dale et al., 2017). This is another field with explosive growth, so much so that some authors have even published on the name of the domain itself (Ricciardi & Milano, 2022). Julian was there from the start, contributing his insights and ideas throughout.CGP and in materio computing may be what Julian is best known for, but these contributions were embedded in a deeper research program of understanding development as a fundamental component of evolving embodied computation. From growing a self-repairing “French-flag” organism (Miller, 2004), to assembling complex structures through Artificial Chemistries (Faulconbridge et al., 2011), to the idea of the “software garden” (Miller, 2018), Julian felt that both growth and evolution are essential concepts in complex systems.His Festschrift Inspired by Nature (Stepney & Adamatzky, 2018) was a (slightly late) 60th birthday present from his many academic colleagues. It includes chapters contributed by a wide range of authors who have built on and been inspired by his many research interests. The text covers evolution and hardware, CGP applications, chemistry, and development. Julian retired in 2016, but he did not stop his research. He used the freedom from the quotidian constraints of an academic job to pursue a new interest. He was bringing together his discoveries in evolution, development, networks, and computation to develop a new neural model to evolve programs that build, or grow, neural networks. His most recent publication on that topic has only just appeared (Miller, 2022).Julian’s retirement also allowed him to spend time with his recently acquired beloved new family. His wife Gabi remembers him thus: “He was a loving and generous-hearted husband, a wise step-father to my three grown adults and much loved Grandpa to our four grand-children. Jules will be sadly missed, but also lovingly remembered by all whose life he touched.” Many further tributes to Julian from his colleagues can be found in the latest SIGEVO newsletter (Ochoa, 2022).

CGP-based Logic Flow: Optimizing Accuracy and Size of Approximate Circuits

Logic synthesis tools face tough challenges when providing algorithms for synthesizing circuits with increased inputs and complexity. Machine learning techniques show high performance in solving specific problems, being an attractive option to improve electronic design tools. We explore Cartesian Genetic Programming (CGP) for logic optimization of exact or approximate Boolean functions in our work. The proposed CGP-based flow receives the expected circuit behavior as a truth-table and either performs the synthesis starting from random circuits or optimizes a circuit description provided in the format of an AND-Inverter Graph. The optimization flow improves solutions found by other techniques, using them for bootstrapping the evolutionary process. We use two metrics to evaluate our CGP-based flow: (i) the number of AIG nodes or (ii) the circuit accuracy. The results obtained showed that the CGP-based flow provided at least 22.6% superior results when considering the trade-off between accuracy and size compared with two other methods that brought the best accuracy and size outcomes, respectively.

An Analysis of the Influence of Noneffective Instructions in Linear Genetic Programming.

Linear Genetic Programming (LGP) represents programs as sequences of instructions and has a Directed Acyclic Graph (DAG) dataflow. The results of instructions are stored in registers that can be used as arguments by other instructions. Instructions that are disconnected from the main part of the program are called noneffective instructions, or structural introns. They also appear in other DAG-based GP approaches like Cartesian Genetic Programming (CGP). This article studies four hypotheses on the role of structural introns: noneffective instructions (1) serve as evolutionary memory, where evolved information is stored and later used in search, (2) preserve population diversity, (3) allow neutral search, where structural introns increase the number of neutral mutations and improve performance, and (4) serve as genetic material to enable program growth. We study different variants of LGP controlling the influence of introns for symbolic regression, classification, and digital circuits problems. We find that there is (1) evolved information in the noneffective instructions that can be reactivated and that (2) structural introns can promote programs with higher effective diversity. However, both effects have no influence on LGP search performance. On the other hand, allowing mutations to not only be applied to effective but also to noneffective instructions (3) increases the rate of neutral mutations and (4) contributes to program growth by making use of the genetic material available as structural introns. This comes along with a significant increase of LGP performance, which makes structural introns important for LGP.

SagTree: Towards efficient mutation in evolutionary circuit approximation

• We thoroughly investigate the performance of various mutation operators for Cartesian Genetic Programming in the context of automated design of complex arithmetic circuits. • We propose a novel mutation operator, so-called SagTree, that is tailored to circuit approximation and combines the classical single active gene mutation with a node deactivation operation. • We perform an extensive experimental comparison with existing mutation operators—our results, grounded on a rigorous statistical evaluation including 39 approximation problems and over 14,000 approximation runs, clearly demonstrate that the SagTree operator significantly outperforms the other operators in CGP-based circuit approximation. • Using the SagTree operator, we obtain complex approximate circuits providing significantly better trade-offs between energy savings and the incurred error compared to existing circuits obtained using both evolutionary and non-evolutionary state-of-the-art techniques. Approximate circuits that trade the chip area for the quality of results play a key role in the development of energy-aware systems. Designing complex approximate circuits is, however, a very difficult and computationally demanding process. Evolutionary approximation—in particular, the method of Cartesian Genetic Programming (CGP)—currently represents one of the most successful approaches for automated circuit approximation. In this paper, we thoroughly investigate mutation operators for CGP with respect to the performance of circuit approximation. We design a novel dedicated operator that combines the classical single active gene mutation with a node deactivation operation (eliminating a part of the circuit forming a tree from an active gate). We show that our new operator significantly outperforms other operators on a wide class of approximation problems (such as 16 bit multipliers and dividers) and thus improves the performance of the state-of-the-art approximation techniques. Our results are grounded on a rigorous statistical evaluation including 39 approximation scenarios and 14,000 runs.

Unlocking approximation for in-memory computing with Cartesian genetic programming and computer algebra for arithmetic circuits

Abstract With ReRAM being a non-volative memory technology, which features low power consumption, high scalability and allows for in-memory computing, it is a promising candidate for future computer architectures. Approximate computing is a design paradigm, which aims at reducing the complexity of hardware by trading off accuracy for area and/or delay. In this article, we introduce approximate computing techniques to in-memory computing. We extend existing compilation techniques for the Programmable Logic in-Memory (PLiM) computer architecture, by adapting state-of-the-art approximate computing techniques for arithmetic circuits. We use Cartesian Genetic Programming for the generation of approximate circuits and evaluate them using a Symbolic Computer Algebra-based technique with respect to error-metrics. In our experiments, we show that we can outperform state-of-the-art handcrafted approximate adder designs.