Canonical quantum algorithms often achieve low execution fidelities on current Noisy Intermediate-Scale Quantum (NISQ) hardware. The standard implementation of Grover's search algorithm, designed for theoretical generality, produces deep, gate-heavy circuits that are susceptible to noise. This paper challenges the "one-size-fits-all" design paradigm by using Grammatical Evolution (GE) to automatically discover hardware-efficient, state-specific quantum circuits. We demonstrate this approach by evolving bespoke circuits for all eight 3-qubit computational basis states and executing them on a 133-qubit IBM Heron quantum processor. To our knowledge, this is the first hardware-validated application of GE for this task. The results indicate significant performance gains: evolved circuits achieve hardware-executed fidelities up to 96.9% (vs. 66.3% baseline) while reducing circuit depth by 82.5–96.6% and gate count by 77.4–94.6% compared to canonical implementations. These findings suggest that automated symbolic search is a viable approach to designing algorithms that can execute on today's NISQ devices.

This study compares Deep Autencoder (AE)-based Normal Behaviour Models (NBM) for anomaly detection in Wind Turbine SCADA data. Using a proprietary industrial dataset, we evaluate performance under real world challenges, like class imbalance and unseen anomalies. We conduct a systematic comparison across unsupervised, semi-supervised, and supervised approaches, and examine the impact of auxiliary loss functions. Our results show that unsupervised Vanilla AEs struggle to separate normal and abnormal data, different from what the NBM literature claims about the effectiveness of unsupervised setups, as we obtain a significant increase in the Area Under the Curve (AUC) via its classification head. We propose an Adversarial Robust AE (ARAE) to improve detection in data-scarce scenarios. In settings with limited abnormal data, ARAE maintains stable recall at a 1:4 abnormal-to-normal ratio, outperforming other models under severe class imbalance. Based on these results, we recommend a Bottleneck architecture for scenarios with abundant labelled data and ARAE for those with scarce abnormal examples.

In this research, we address bearing fault classification by evaluating three neural network models: 1D Convolutional Neural Network (1D-CNN), CNN-Visual Geometry Group (CNN-VGG), and Long Short-Term Memory (LSTM). Utilizing vibration data, our approach incorporates data augmentation to address the limited availability of fault class data. A significant aspect of our methodology is the application of neural architecture search (NAS), which automates the evolution of network architectures, including hyperparameter tuning, significantly enhancing model training. Our use of early stopping strategies effectively prevents overfitting, ensuring robust model generalization. The results highlight the potential of integrating advanced machine learning models with NAS in bearing fault classification and suggest possibilities for further improvements, particularly in model differentiation for specific fault classes.

With the growing popularity of machine learning (ML), regression problems in many domains are becoming increasingly high-dimensional. Identifying relevant features from a high-dimensional dataset still remains a significant challenge for building highly accurate machine learning models.

Bloat, a well-known phenomenon in Evolutionary Computation, often slows down evolution and complicates the task of interpreting the results. We propose Lexi2, a new selection and bloat-control method, which extends the popular lexicase selection method, by including a tie-breaking step which considers attributes related to the size of the individuals. This new step applies lexicographic parsimony pressure during the selection process and is able to reduce the number of random choices performed by lexicase selection (which happen when more than a single individual correctly solve the selected training cases).

Deep learning is a well suited approach to successfully address image processing and there are several Neural Networks architectures proposed on this research field, one interesting example is the U-net architecture and and its variants. This work proposes to automatically find the best architecture combination from a set of the current most relevant U-net architectures by using a genetic algorithm (GA) applied to solve the Retinal Blood Vessel Segmentation (RVS), which it is relevant to diagnose and cure blindness in diabetes patients. Interestingly, the experimental results show that avoiding human-bias in the design, GA finds novel combinations of U-net architectures, which at first sight seems to be complex but it turns out to be smaller, reaching competitive performance than the manually designed architectures and reducing considerably the computational effort to evolve them.

The advent of the Covid-19 pandemic has resulted in a global crisis making the health systems vulnerable, challenging the research community to find novel approaches to facilitate early detection of infections. This open-up a window of opportunity to exploit machine learning and artificial intelligence techniques to address some of the issues related to this disease. In this work, we address the classification of ten SARS-CoV-2 protein sequences related to Covid-19 using k-mer frequency as features and considering two objectives; classification performance and feature selection. The first set of experiments considered the objectives one at the time, four techniques were used for the feature selection and twelve well known machine learning methods, where three are neural network based for the classification. The second set of experiments considered a multi-objective approach where we tested a well known multi-objective approach Non-dominated Sorting Genetic Algorithm II (NSGA-II), and the Multi-dimensional Archive of Phenotypic Elites (MAP-Elites), which considers quality+diversity containers to guide the search through elite solutions. The experimental results shows that ResNet and PCA is the best combination using single objectives. Whereas, for the mulit-classification, NSGA-II outperforms ME with two out of three classifiers, while ME gets competitive results bringing more diverse set of solutions.

Exploiting patterns within a solution or reusing certain functionality is often necessary to solve certain problems. This paper proposes a new method for identifying useful modules. Modules are only considered if they are prevalent in the population and they are seen to have a positive effect on an individual's fitness. This is achieved by finding the covariance of an individual's fitness with the presence of a particular subtree in the overall expression.While there are many successful systems that dynamically add modules during Genetic Programming (GP) runs, doing so is not trivial for Grammatical Evolution (GE), due to the fact that it employs a mapping process to produce individuals from binary strings, which makes it difficult to dynamically change the mapping process during a run.We adopt a multi-run approach which only has a single stage of module addition to mitigate the problems associated with continuously adding newly found functionality to a grammar. Based on the well-known Price Equation, our system explores the covariance between traits to identify useful modules, which are added to the grammar, before the system is restarted. Grammar Augmentation through Module Encapsulation (GAME) was tested on seven problems from three different domains and was observed to significantly improve the performance on 3 problems and never showing harmful effects on any problem. GAME found the best individual in 6 of the 7 experiments.

We present Pyramid, a Hierarchical Genetic Algorithm that decomposes problems by first tackling simpler versions of them, before automatically scaling up to more difficult versions while also reducing the population size. Pyramid takes its name from the architectural phenomenon of the Mayan pyramid in Chichen-Itza, which, although constructed from the bottom up, operates in a top down manner through interactions with the sun. This gives a two stage approach: initially we create our pyramid of experiments, with the most complex fitness functions at the bottom, and with increasingly more simplified/decomposed version as we move up through the pyramid. Runs start at the top of the pyramid and populations descend through it, decreasing in size, being exposed to increasingly complex fitness functions, until, at the bottom layer, we have small populations with the original full fitness function. We conduct experiments comparing the performance of Pyramid for a suite of difficult unimodal and multimodal functions. The experimental results show that in three of four cases, Pyramid achieves the same or better fitness scores as standard algorithms, with substantially fewer evaluations, and that in two cases it actually performs statistically significantly better.

Heuristic-based search techniques have been increasingly used to automate different aspects of software testing. Several studies suggest that variable interdependencies may exist in branching conditions of real-life programs, and these dependencies result in the need for highly precise data values (such as of the form i=j=k) for code coverage analysis. This requirement makes it very difficult for Genetic Algorithm (GA)-based approach to successfully search for the required test data from vast search spaces of real-life programs.

Time series forecasting is a technique that predicts future values using time as one of the dimensions. The learning process is strongly controlled by fine-tuning of various hyperparameters which is often resource extensive and requires domain knowledge. This research work focuses on automatically evolving suitable hyperparameters of time series for level, trend and seasonality components using Grammatical Evolution. The proposed Grammatical Evolution Time Series framework can accept datasets from various domains and select the appropriate parameter values based on the nature of dataset. The forecasted results are compared with a traditional grid search algorithm on the basis of error metric, efficiency and scalability.

In Grammatical Evolution (GE) individuals occupy more space than required, that is, the Actual Length of the individuals is longer than their Effective Length. This has major implications for scaling GE to complex problems that demand larger populations and complex individuals. We show how these two lengths vary for different sizes of population, demonstrating that Effective Length is relatively independent of population size, but that the Actual Length is proportional to it. We introduce Grammatical Evolution Memory Optimization (GEMO), a two-stage evolutionary system that uses a multi-objective approach to identify the optimal, or at least, near-optimal, genome length for the problem being examined. It uses a single run with a multi-objective fitness function defined to minimize the error for the problem being tackled along with maximizing the ratio of Effective to Actual Genome Length leading to better utilization of memory and hence, computational speedup. Then, in Stage 2, standard GE runs are performed restricting the genome length to the length obtained in Stage 1. We demonstrate this technique on different problem domains and show that in all cases, GEMO produces individuals with the same fitness as standard GE but significantly improves memory usage and reduces computation time.

Blood vessel extraction in digital retinal images is an important step in medical image analysis for abnormality detection and also obtaining good retinopathy diabetic diagnosis; this is often referred to as the Retinal Blood Vessel Segmentation task and current state-of-the-art approaches all use some form of neural networks. Designing neural network architecture and selecting appropriate hyper-parameters for a specific task is challenging. In recent works, increasingly more complex models are starting to appear, but in this work, we present a simple and small model with a very low number of parameters with good performance compared with the state of the art algorithms. In particular, we choose a standard Genetic Algorithm (GA) for selecting the parameters of the model and we use an expert-designed U-net based model, which has become a very popular tool in image segmentation problems. Experimental results show that GA is able to find a much shorter architecture and acceptable accuracy compared to the U-net manually designed. This finding puts on the right track to be able in the future to implement these models in portable applications

The importance of environmental fluctuations in the evolution of living organisms by natural selection has been widely noted by biologists and linked to many important characteristics of life such as modularity, plasticity, genotype size, mutation rate, learning, or epigenetic adaptations.In artificial-life simulations, however, environmental fluctuations are usually seen as a nuisance rather than an essential characteristic of evolution.HetCA is a heterogeneous cellular automata characterized by its ability to generate open-ended long-term evolution and "evolutionary progress".In this paper, we propose to measure the impact of different types of environmental fluctuations in HetCA.Our results indicate that environmental changes induce mechanisms analogous to epigenetic adaptation or multilevel selection.This is particularly prevalent in two of the tested fluctuation schemes, which involve a round-robin inhibition of certain cell types, where phenotypic selection seems to occur.

Writing correct and efficient parallel programs is an unavoidable challenge; the challenge becomes arduous with lock-free programming. This paper presents an automated approach, Automatic Lock-free Programming (ALP) that avoids the programming difficulties via locks for an average programmer. ALP synthesizes parallel lock-free recursive programs that are directly compilable on multi-core processors. ALP attains the dual objective of evolving parallel lock-free programs and optimizing their performance. These programs perform (in terms of execution time) significantly better than that of the parallel programs with locks, while they are competitive with that of the human developed programs.

Wave is a novel form of semantic genetic programming which operates by optimising the residual errors of a succession of short genetic programming runs, and then producing a cumulative solution. These short genetic programming runs are called periods, and they have heterogeneous parameters. In this paper we leverage the potential of Wave's heterogeneity to simulate a dynamic evolutionary environment by incorporating self adaptive parameters together with an innovative approach to population renewal. We conduct an empirical study comparing this new approach with multiple linear regression~(MLR) as well as several evolutionary computation~(EC) methods including the well known geometric semantic genetic programming~(GSGP) together with several other optimised Wave techniques. The results of our investigation show that the dynamic Wave algorithm delivers consistently equal or better performance than Standard GP (both with or without linear scaling), achieves testing fitness equal or better than multiple linear regression, and performs significantly better than GSGP on five of the six problems studied.

Although Evolutionary Computation (EC) has been used with considerable success to evolve computer programs, the majority of this work has targeted the production of serial code. Recent work with Grammatical Evolution (GE) produced Multi-core Grammatical Evolution (MCGE-II), a system that natively produces parallel code, including the ability to execute recursive calls in parallel. This paper extends this work by including practical constraints into the grammars and fitness functions, such as increased control over the level of parallelism for each individual. These changes execute the best-of-generation programs faster than the original MCGE-II with an average factor of 8.13 across a selection of hard problems from the literature. We analyze the time complexity of these programs and identify avoiding excessive parallelism as a key for further performance scaling. We amend the grammars to evolve a mix of serial and parallel code, which spawns only as many threads as is efficient given the underlying OS and hardware; this speeds up execution by a factor of 9.97.

We present an automated, end-to-end approach for Stage~1 breast cancer detection. The first phase of our proposed work-flow takes individual digital mammograms as input and outputs several smaller sub-images from which the background has been removed. Next, we extract a set of features which capture textural information from the segmented images.

For some time, Genetic Programming research has lagged behind the wider Machine Learning community in the study of generalisation, where the decomposition of generalisation error into bias and variance components is well understood. However, recent Genetic Programming contributions focusing on complexity, size and bloat as they relate to over-fitting have opened up some interesting avenues of research. In this paper, we carry out a simple empirical study on five binary classification problems. The study is designed to discover what effects may be observed when program size and complexity are varied in combination, with the objective of gaining a better understanding of relationships which may exist between solution size, operator complexity and variance error. The results of the study indicate that the simplest configuration, in terms of operator complexity, consistently results in the best average performance, and in many cases, the result is significantly better. We further demonstrate that the best results are achieved when this minimum complexity set-up is combined with a less than parsimonious permissible size.

Genetic programming (GP) coarsely models natural evolution to evolve computer programs. Unlike in nature, where individuals can often improve their fitness through lifetime experience, the fitness of GP individuals generally does not change during their lifetime, and there is usually no opportunity to pass on acquired knowledge. This paper introduces the Chameleon system to address this discrepancy and augment GP with lifetime learning by adding a simple local search that operates by tuning the internal nodes of individuals. Although not the first attempt to combine local search with GP, its simplicity means that it is easy to understand and cheap to implement. A simple cache is added which leverages the local search to reduce the tuning cost to a small fraction of the expected cost, and we provide a theoretical upper limit on the maximum tuning expense given the average tree size of the population and show that this limit grows very conservatively as the average tree size of the population increases. We show that Chameleon uses available genetic material more efficiently by exploring more actively than with standard GP, and demonstrate that not only does Chameleon outperform standard GP (on both training and test data) over a number of symbolic regression type problems, it does so by producing smaller individuals and it works harmoniously with two other well-known extensions to GP, namely, linear scaling and a diversity-promoting tournament selection method.

NAND Flash memory is a multi-billion dollar industry which is projected to continue to show significant growth until at least 2017. Devices such as smart-phones, tablets and Solid State Drives use NAND Flash since it has numerous advantages over Hard Disk Drives including better performance, lower power consumption, and lower weight. However, storage locations within Flash devices have a limited working lifetime, as they slowly degrade through use, eventually becoming unreliable and failing. The number of times a location can be programmed is termed its endurance, and can vary significantly, even between locations within the same device. There is currently no technique available to predict endurance, resulting in manufacturers placing extremely conservative specifications on their Flash devices. We perform symbolic regression using Genetic Programming to estimate the endurance of storage locations, based only on the duration of program and erase operations recorded from them. We show that the quality of estimations for a device can be refined and improved as the device continues to be used, and investigate a number of different approaches to deal with the significant variations in the endurance of storage locations. Results show this technique's huge potential for real-world application.

In this work we study open-ended evolution through the analysis of a new model, HetCA, for "heterogeneous cellular automata". Striving for simplicity, HetCA is based on classical two-dimensional CA, but differs from them in several key ways: cells include properties of "age", "decay", and "quiescence"; cells utilize a heterogeneous transition function, one inspired by genetic programming; and there exists a notion of genetic transfer between adjacent cells. The cumulative effect of these changes is the creation of an evolving ecosystem of competing cell colonies. To evaluate the results of our new model, we define a measure of phenotypic diversity on the space of cellular automata. Via this measure, we contrast HetCA to several controls known for their emergent behaviours---homogeneous CA and the Game of Life---and several variants of our model. This analysis demonstrates that HetCA has a capacity for long-term phenotypic dynamics not readily achieved in other models. Runs exceeding one million time steps do not exhibit stagnation or even cyclic behaviour. Further, we show that the design choices are well motivated, as the exclusion of any one of them disrupts the long-term dynamics.

This paper explores a range of class boundary determination techniques that can be used to improve performance of Genetic Programming (GP) on binary classification tasks. These techniques involve selecting an individualised boundary threshold in order to reduce implicit bias that may be introduced through employing arbitrarily chosen values. Individuals that can chose their own boundaries and the manner in which they are applied, are freed from having to learn to force their outputs into a particular range or polarity and can instead concentrate their efforts on seeking a problem solution.

This paper introduces an approach to evolving computer programs using an Attribute Grammar (AG) extension of Grammatical Evolution (GE) to eliminate ineffective pieces of code with the help of context-sensitive information.

The aim of a modern Building Automation System (BAS) is to enhance interactive control strategies for energy efficiency and user comfort. In this context, we develop a novel control algorithm that uses a stochastic building occupancy model to improve mean energy efficiency while minimizing expected discomfort. We compare by simulation our Stochastic Model Predictive Control (SMPC) strategy to the standard heating control method to empirically demonstrate a 4.3% reduction in energy use and 38.3% reduction in expected discomfort.

This paper proposes to improve the performance of Genetic Programming (GP) over unseen data by minimizing the variance of the output values of evolving models alongwith reducing error on the training data. Variance is a well understood, simple and inexpensive statistical measure; it is easy to integrate into a GP implementation and can be computed over arbitrary input values even when the target output is not known.

This paper describes a technique which can be used with Genetic Programming (GP) to reduce implicit bias in binary classification tasks. Arbitrarily chosen class boundaries can introduce bias, but if individuals can choose their own boundaries, tailored to their function set, then their outputs are automatically scaled into a suitable range. These boundaries evolve over time as the individuals adapt to the data. Our system calculates the Evolved Class Boundary(ECB) for each individual in every generation, with the twin aims of reducing training times and improving test fitness. The method is tested on three benchmark binary classification data sets from the medical domain.

Typically, an individual in Genetic Programming (GP) can not make the most of its genetic inheritance. Once it is mapped, its fitness is immediately evaluated and it survives only until the genetic operators and its competitors eliminate it. Thus, the key to survival is to be born strong.

Evolutionary algorithms based on probabilistic modeling is a growing research field. Hybrids that borrow ideas from the field of classification were introduced. We extend such hybrids, and evaluate four strategies for truncation of an over-sized population of samples. The strategies are evaluated over a number of difficult problems from the literature, among them, a hierarchical 256-bit HIFF problem. We show that over-sampling in conjunction with a truncation strategy can guide the search without increasing the number of performed fitness evaluations per generation, and that a truncation strategy which inverses the sampling pressure can, fitness-wise, perform significantly better than regular sampling.

We have examined the application of genetic Algorithms to noisy fitness functions and consider the accepted wisdom of sampling, or multiple evaluations of individuals, as a mechanism for identifying true performance. Given a large (> 10%) amount of noise, a standard GA of surprisingly modest population size outperforms a GA using sampling, when compared on fitness versus evaluations. We also document a detrimental phenomenon we term the Glass Ceiling, which is when individuals of high fitness become confused with individuals of perfect fitness by the GA. We pinpoint the precise conditions that create this effect.

Several algorithms within the field of Evolutionary Computation have been proposed that effectively turn optimisation problems into supervised learning tasks. Typically such hybrid algorithms partition their populations into three subsets, high performing, low performing and mediocre, where the subset containing mediocre candidates is discarded from the phase of model construction. In this paper we will empirically compare this traditional partitioning scheme against two alternative schemes on a range of difficult problems from the literature. The experiments will show that at small population sizes, using the whole population is often a better approach than the traditional partitioning scheme, but partitioning around the midpoint and ignoring candidates at the extremes, is often even better.

Hereditary Repulsion (HR) is a selection method coupled with a fitness constraint that substantially improves the performance and consistency of evolutionary algorithms. This also manifests as improved generalisation in the evolved GP expressions. We examine the behaviour of HR on the difficult Parity 5 problem using a population size of only 24 individuals. The negative effects of convergence are amplified under these circumstances and we progress through a series of insights and experiments which dramatically improve the consistency of the algorithm, resulting in a 70% success rate with the same small population. By contrast, a steady state GP system using a population of 5000 only had a success rate of 8%. We then confirm the effectiveness of these results in a number of arbitrary problem domains.

Voice over IP (VoIP) speech quality estimation is crucial to providing optimal Quality of Service (QoS). This paper seeks to provide improved speech quality estimation models with better prediction accuracy by considering a richer set of input features than the current International Telecommunications Union-Telecommunication (ITU-T) recommendations. It addresses a transitional phase, where wideband (WB) networks are becoming available. However, they have to co-exist with the existing narrowband (NB) setups for the time being. Quality estimation becomes a challenge in such a mixed context. The ITU-T recommendation (termed E-Model) has recently been extended to deal with the mixed context. However, it evaluates the speech degradation in the WB scenario based solely on codec related distortions (only a subset of factors affecting the speech quality on a VoIP network). The extension is derived out of speech signals evaluated by human subjects: an expensive and difficult to reproduce exercise. This paper innovates by considering a number of other network distortion types as well to produce generalised models that predict the quality degradation to a higher accuracy. To this end, an extensive set of speech samples is subjected to a wide variety of distortions. The degraded signals are evaluated by the currently best available algorithmic approximation of human evaluation of speech to produce quality scores. Using the distortions as the input features and targeting the quality scores, we employ Genetic Programming to produce parsimonious models that show considerable prediction gain compared to the E-Model. As against some existing approaches, where the models are tailored to various telephony codecs, the evolved models generalise across a variety of modern codecs.

This paper describes a novel Genetic Algorithm (GA) approach to scheduling. Although the particular problems examined are all multi-processor scheduling types it can, because the algorithm takes a DAG (Directed Acyclic Graph) as input, be applied to any scheduling problem represented by a DAG.

This paper proposes a novel approach to quantifying the quality degradation of Voice over IP (VoIP) telephony in the presence of codec and network-related impairments. This approach differs from the baisc ITU-T E-Model for VoIP quality estimation in that it addresses mixed narrowband/wideband scenarios. It makes novel use of instrumental models and symbolic regression via Genetic Programming (GP) to enable the evolution of degradation models from a modest set of initial parameters. Here, a two-step approach has been used. First, values of impairment factors are derived using WB-PESQ as a reference model. Secondly, a GP based symbolic regression approach has been utilized to automatically evolve the functional form of equipment impairment factors from a set of variables. Very few a priori assumptions are made about the model structure. The effectiveness of the approach is demonstrated by a number of generated models which compare favorably with WB-PESQ and outperform the traditional E-Model in terms of prediction accuracy when compared using WB-PESQ. A significant advantage of the approach is that new models are easily generated to account for continuing evolution of the VoIP standards.

Estimation of distribution algorithms (EDA) is an active area of research within the field of evolutionary algorithms. While EDAs have shown great promise on difficult problems with strong epistasis between genes, such as hierarchical and deceptive problems, they have not been a choice for non-stationary problems where the target solution changes over time. This work aims to explore the diversity within the population of an EDA using a supervised classifier. We introduce a technique, sampling-mutation, that can help increase the useful diversity within the population. We show that sampling-mutation increases the performance of an EDA on a non-stationary problem and a hierarchical problem.

We describe the "bottom-up" construction of a system which aims to build models of human musicalpreferences with strong predictive power. We use Grammatical Evolution to construct models from toydatasets which mimic real-world user-generated data. These models will ultimately substitute for the subjective fitness functions that human users employ during Interactive Evolution of melodies.

This paper introduces a new type of mutation, Context-Aware Mutation, which is inspired by the recently introduced context-aware crossover. Context-Aware mutation operates by replacing existing sub-trees with modules from a previously construct repository of possibly useful sub-trees.We describe an algorithmic way to produce the repository from an initial, exploratory run and test various GP set ups for producing the repository. The results show that when the exploratory run uses context-aware crossover and the main run uses context-aware mutation, not only is the final result significantly better, the overall cost of the runs in terms of individuals evaluated is significantly lower.

Crossover in Genetic Programming is mostly a destructive operator, generally producing children worse than the parents and occasionally producing those who are better. A recently introduced operator, Context-Aware Crossover, which implicitly discovers the best possible crossover site for a subtree has been shown to consistently attain higher fitnesses while processing fewer individuals.It has been observed that context-aware crossover is similar to Brood Crossover in that multiple children are produced during each crossover event. This paper performs a thorough analysis of these crossover operators and compares the performance of the two and demonstrates that, although they do work similarly, context-aware crossover performs a far better sampling of the search space and thus performs much better.We also demonstrate that context-aware crossover benefits from a speed up of almost an order of magnitude when using a simple and very small cache, which is over two orders of magnitute smaller than caches typically used.

Run Transferable Libraries (RTL) is an extension for GP where individualsin a population choose functions from an external library of ADF-likefunctions rather than from a set of standard GP functions. All previous work done with RTL provided a predefined function set. Thiswork investigates mechanisms by which the library can be seeded with domainrelevent functionality. .

This paper describes the application of evolutionary search to the problem of Flash memory wear-out. The operating parameters of Flash memory are notoriously difficult to determine, as the optimal values vary from batch to batch. These parameters are usually established by an expensive, once off process of manual destructive testing at design time. Testing on individual batches is normally not feasible. We establish the viability of a platform that performs destructive experimentation on hard silicon, using a Genetic Algorithm to automatically discover optimal operating parameter settings. The results demonstrate a minimum average life extension of between 250% and 350% over the factory set read write and erase conditions with a maximum life extension exhibited of 700% for cells within the same device. It was necessary to build specialized hardware to perform the repetitive testing required by the GA, here we describe this hardware and demonstrate how the lessons learned in this pilot study will allow us to proceed with a more complex parallel evaluation platform, which will facilitate a larger problem space, larger population size and diversity of search techniques, facilitating the near no cost life extension of a split-gate Flash memory device.

This paper investigates the effects of the mapping process employed by the GAuGE system on standard genetic operators. It is shown that the application of that mapping process transforms these operators into suitable sequencing searching tools. A practical application is analysed, and its results compared with a standard genetic algorithm, using the same operators. Results and analysis highlight the suitability of GAuGE and its operators, for this class of problems.

This paper describes the use of a recently introduced crossover operator for GP, context-aware crossover. Given a randomly selected subtree from one parent, context-aware crossover will always find the best location to place the subtree in the other parent.We examine the performance of GP when context-aware crossover is used as an extra crossover operator, and show that standard crossover is far more destructive, and that performance is better when only context-aware crossover is used.There is still a place for standard crossover, however, and results suggest that using standard crossover in the initial part of the run and then switching to context-aware crossover yields the best performance.We show that, across a range of standard GP benchmark problems, context-aware crossover produces a higher best fitness as well as a higher mean fitness, and even manages to solve the 11-bit multiplexer problem without ADFs. Furthermore, the individuals produced this way are much smaller than standard GP, and far fewer individual evaluations are required, so GP achieves a higher fitness by evaluating fewer and smaller individuals.

This paper presents a new migration strategy that improves the overall quality of solutions in a distributed genetic algorithm (DGA) involving a number of concurrently evolving populations. The idea behind this improvement is to incorporate a diversity guided selection mechanism that selects a diverse set of individuals for migration from the evolving populations. To accompany this selection mechanism an alternative replacement policy which replaces individuals that have more than one of their copies present in the population (clones) is also investigated. This increases diversity within a population and reduces premature convergence. Results show that it leads to a better performance when compared with the send-best-replace-worst strategy.

In this paper we introduce a cooperative revolutionary algorithm based on the ideas of endosymbiosis. We compare it to a generational GA on two deceptive and decomposable problems and show that it has better scaling properties as the problem size increases. We then analyse what effect crossover and parasite mutation has on its performance and conclude that a high parasite mutation rate is preferred over a lower rate and that crossover has no, or a very small, effect on its performance.

We propose a new methodology to look at the fitness contributions (semantics) of different schemata in Genetic Programming (GP). We hypothesize that the significance of a schema can be evaluated by calculating its fitness contribution to the total fitness of the trees that contain it, and use our methodology to test this hypothesis.It is shown that this method can also be used to identify schemata that are important in terms of both individual runs and individual problems (that is, schema that will be important across many runs on a particular problem).The usefulness of this study to existing schema theories and its effective use in the detection of introns, in the identification of potentially useful modular functions are also discussed in this paper.

We present grammatical evolution, an evolutionary algorithm that can evolve complete programs in an arbitrary language using a variable-length binary string. The binary genome determines which production rules in a Backus-Naur form grammar definition are used in a genotype-to-phenotype mapping process to a program. We demonstrate how expressions and programs of arbitrary complexity may be evolved and compare its performance to genetic programming.