Fact-checked by Grok 2 weeks ago

Evolutionary programming

Evolutionary programming (EP) is a and paradigm within the field of that simulates the process of to generate adaptive solutions for complex problems, primarily through mechanisms of and selection without the use of genetic crossover. Originating in the early , EP was pioneered by J. Fogel as an approach to , focusing on the evolution of finite state machines to perform tasks such as prediction and . This method emphasizes behavioral adaptation over explicit genetic representation, allowing populations of candidate solutions to iteratively improve their in response to environmental pressures. The foundational work on EP traces back to Fogel's 1962 exploration of autonomous automata and was formalized in his 1966 book Artificial Intelligence through Simulated Evolution, co-authored with A.J. Owens and M.J. Walsh, where experiments demonstrated the evolution of predictive models outperforming fixed strategies. In the decades following, David B. Fogel, Lawrence's son, extended EP to broader applications, including continuous parameter optimization and the training of neural networks, introducing self-adaptive mutation operators that dynamically adjust variation rates based on performance feedback. These advancements positioned EP as a robust alternative to other evolutionary algorithms, such as genetic algorithms (which rely heavily on crossover) and evolution strategies (which focus on real-valued parameters), by prioritizing domain-independent behavioral evolution. Key features of EP include a population-based search where each represents a potential solution—often encoded as a or real-valued vector—that undergoes Gaussian mutation to produce offspring, followed by tournament selection to retain the fittest members for the next generation. This process enables EP to handle non-differentiable, noisy, or objective functions effectively, making it suitable for design, scheduling, and biological modeling. Notable applications have included evolving controllers for robotic systems, optimizing designs, and developing strategies for game playing, such as chess programs that learn without explicit encoding. Despite its strengths in self-adaptation, EP's computational intensity for large-scale problems has led to hybrid integrations with local search techniques in contemporary research.

Overview

Definition and core principles

Evolutionary programming (EP) is a , -based optimization technique that simulates the process of biological to search for solutions to complex problems. Developed as a method for generating , EP evolves a of candidate solutions—initially represented as finite state machines (FSMs)—across generations by applying variation operators and selection pressures to minimize or maximize a predefined fitness function. Unlike traditional optimization approaches, EP treats the solutions as adaptive organisms that improve their performance in response to an environmental challenge, such as predicting sequential data. At its core, EP draws on Darwinian principles of variation and , but it uniquely emphasizes phenotypic evolution, focusing on behavioral outcomes rather than explicit genetic structures. Variation is achieved primarily through , which introduces random changes to individual solutions to generate offspring, while recombination (crossover) is absent in the classical formulation to maintain a direct linkage between parental and offspring behaviors. Selection then favors individuals with superior , typically measured by their ability to perform well in the target task, ensuring that the population progressively adapts over iterations. This process continues until a satisfactory solution is found or is reached, mimicking how natural populations evolve without relying on mechanisms. Central to EP are several key concepts that define its operational framework. A consists of multiple individuals, each encoding a potential solution in a problem-specific , such as the state transitions and outputs of an FSM. evaluation quantifies an individual's effectiveness in the domain, for instance, by scoring the accuracy of predictions against observed data sequences. Generational progression involves creating a new population from the fittest parents via , followed by competitive selection to retain the most promising candidates, thereby guiding the search toward optimal or near-optimal behaviors. The historical motivation for EP stemmed from efforts to evolve intelligent behaviors in artificial systems, particularly by treating programs as living entities that adapt to dynamic environments. Originally applied to the evolution of FSMs for time-series tasks, EP viewed these machines as organisms whose structures and parameters self-adapt through evolutionary processes, enabling them to forecast future events based on past observations without human-designed rules. This approach highlighted EP's role within the broader field of as a for behavioral optimization.

Relation to other evolutionary algorithms

Evolutionary programming (EP) is one of the four primary paradigms within the broader field of (EC), alongside genetic algorithms (GA), evolution strategies (ES), and (GP). These paradigms collectively employ population-based optimization techniques inspired by biological , featuring iterative processes of variation and selection to evolve solutions toward problem optima. While sharing this foundational structure, they diverge significantly in solution representations and variation mechanisms, allowing EP to emphasize behavioral adaptation in domains like finite state machines and real-valued optimization. A core distinction of EP lies in its reliance on as the sole variation , applied to real-valued or behavioral representations without crossover, in contrast to GA's binary string encodings and heavy use of crossover to simulate . EP also differs from , which targets continuous optimization through self-adaptive rates and often includes recombination, positioning more toward numerical fine-tuning. Furthermore, EP contrasts with GP's tree-based evolution of computer programs, where crossover operates on rather than EP's focus on phenotypic . Despite these differences, all four paradigms incorporate selection to favor superior individuals for propagation, ensuring survival of the fittest across generations. EP's representational flexibility particularly bridges GA's discrete search spaces and ES's continuous ones, enabling versatile applications that draw from multiple influences. The nomenclature surrounding these techniques evolved in the 1990s, shifting from isolated terms like "evolutionary programming" to the unified "evolutionary computation" framework, which encouraged hybridization and cross-pollination of operators and representations among EP, GA, ES, and GP.

History

Origins in the 1960s

Evolutionary programming (EP) was invented in 1960 by Lawrence J. Fogel while serving as a program director at the . Motivated by the limitations of early approaches that relied on explicit rule-based programming, Fogel sought to simulate the evolutionary to develop machine intelligence, emphasizing adaptation through variation and selection rather than predefined instructions. This work positioned EP as a response to the challenges in modeling intelligent behavior in uncertain environments, drawing inspiration from biological evolution to evolve predictive capabilities autonomously. In its initial formulation, EP represented candidate solutions as finite state machines (FSMs), treating these machines as "organisms" in a population. The FSMs were tasked with predicting the next symbol in sequences of binary symbols, simulating simple language prediction or pattern recognition. Evolution proceeded by mutating FSM transitions—such as altering outputs, adding or deleting states—to generate offspring, with selection based on a payoff function measuring predictive accuracy, often the squared error between predicted and actual symbols. This behavioral focus, rather than genetic encoding, distinguished EP from contemporaneous methods and aimed to evolve adaptive strategies without human-designed rules. Fogel's foundational ideas were first detailed in his 1962 paper "Autonomous Automata," published in Industrial Research, which introduced the concept of self-organizing machines through simulated evolution. This was expanded in his 1964 Ph.D. dissertation, "On the Organization of Intellect," at the , where he formalized the evolutionary process for intellect simulation. The seminal 1966 book, Artificial Intelligence through Simulated Evolution, co-authored with Alvin J. Owens and Michael J. Walsh, synthesized these efforts and provided comprehensive experiments demonstrating EP's viability. Early applications of EP in the 1960s focused on evolving FSMs for time-series prediction and , such as binary sequences to model basic linguistic structures. Additional demonstrations included game-playing strategies, like simulating air-to-air combat tactics through coevolved opponent behaviors. These efforts culminated in the founding of Decision Science, Inc. in 1965 by Fogel, Owens, and Walsh, the first company dedicated to commercial applications of for real-world problems in prediction and control.

Developments from the 1980s onward

In the 1980s, David Fogel revived and extended (EP) beyond its original focus on finite state machines, applying it to generalized optimization problems using representations such as real-valued vectors, permutations, and matrices. This shift emphasized EP's versatility for numerical and , building on Lawrence Fogel's foundational ideas while adapting the method to broader computational challenges. In 1988, David Fogel introduced tournament selection as a probabilistic mechanism to replace the original fitness-proportional selection, enabling softer competition among solutions and improving convergence on problems like the traveling salesman. A key milestone in the early 1990s was the incorporation of self-adaptation, where strategy parameters for controlling mutation rates were evolved alongside the primary solutions, allowing the algorithm to automatically adjust variation intensity without external tuning. This innovation, detailed in papers by Fogel et al. in 1991 and 1992, introduced meta-evolutionary programming, enhancing EP's robustness for dynamic environments by permitting correlations between object and strategy variables. During the 1990s, EP integrated more deeply into the broader field of , gaining traction through theoretical advancements and applications in areas such as control systems and . David Fogel's 1995 book, Evolutionary Computation: Toward a New Philosophy of Machine , synthesized these developments, arguing for EP's role in advancing machine intelligence through simulated and highlighting its philosophical implications for computation. By the decade's end, enhancements like adaptive strategies further improved EP's efficiency on benchmark functions. In the 2000s, EP saw commercialization through ventures like , Inc., founded by David Fogel in 1993 to apply evolutionary methods to real-world engineering problems, including and applications. From the to 2025, developments emphasized via , with distributed implementations on high-performance clusters enabling EP to handle large-scale optimization, such as in multi-objective problems and scenarios, though it remained less dominant than hybrid approaches with . Integration with contemporary , particularly neural networks, was limited until the 2020s, when evolutionary techniques began supporting for tasks like architecture search and plasticity adaptation.

Algorithmic components

Solution representation

In classical evolutionary programming, candidate solutions are represented as finite state machines (FSMs), each comprising a finite set of states, transitions between those states triggered by input , and outputs associated with each state or transition. These FSMs form complete diagrams for each individual in the population, enabling the evolution of predictive behaviors for tasks such as sequence from non-stationary . Modern evolutionary programming employs more adaptable representations tailored to the problem domain, with a prevalent shift to real-valued vectors for , where each individual is encoded as a D-dimensional array of real numbers denoting adjustable parameters. strings serve as an alternative for challenges, while other formats like permutations or variable-length lists accommodate specific needs such as scheduling or evolving structures. Central to these encodings in evolutionary programming is of mapping "genotypes"—the internal representations such as FSM diagrams or vectors—to "phenotypes," the observable behaviors or al outputs they produce, without imposing a rigid to allow domain-specific . For instance, in optimization for minimization, real-valued vectors directly embody the , yielding phenotypes as the evaluated values. This representational approach emphasizes behavioral efficacy over syntactic structure, facilitating the direct of performance in diverse applications like policy design, where parameter vectors capture essential dynamics without extraneous constraints.

Mutation and variation operators

In evolutionary programming (EP), acts as the primary mechanism for introducing variation into the , supplanting crossover operators common in other evolutionary algorithms; it generates by applying small, random perturbations to parent solutions, enabling gradual exploration of the space. Classical EP, as originally formulated by Fogel, targets finite machines (FSMs) representing predictive models, where involves probabilistic alterations to machine components such as transitions or outputs to evolve behavioral capabilities. Specifically, each offspring is produced from a parent by applying a Poisson-distributed number of (with mean rate typically 3.0), where each equally likely selects one of five operations: adding a (up to a maximum of 25 ), deleting a (maintaining at least 3 ), changing the initial , altering an output symbol, or modifying a next-state transition, with affected elements chosen uniformly at random. Early implementations, including Gaussian perturbations for continuous aspects, emphasized these structural changes to adapt FSMs for tasks like time-series prediction. Modern EP shifts focus to real-valued representations for numerical optimization, employing Gaussian on vectors to produce . For an individual with object variables \mathbf{x} = (x_1, \dots, x_n) and associated strategy \boldsymbol{\sigma} = (\sigma_1, \dots, \sigma_n), each component is mutated independently as x_i' = x_i + N(0, \sigma_i), where N(0, \sigma_i) denotes a normally distributed random deviate with mean 0 and standard deviation \sigma_i, controlling the magnitude for that . This approach, without recombination, relies on alone to navigate continuous search spaces, often with fixed population sizes of 50–100 individuals to balance computational efficiency and diversity. Boundary handling for constrained domains typically involves (e.g., values beyond limits) or repair to ensure feasibility. To enhance adaptability, modern EP incorporates self-adaptation of the strategy parameters \boldsymbol{\sigma}, allowing to evolve strengths alongside solutions. The update for each \sigma_i follows a log-normal scheme: \sigma_i' = \sigma_i \exp\left( \tau' N(0,1) + \tau N_i(0,1) \right), where N(0,1) and N_i(0,1) are independent standard random variables (with N(0,1) shared globally across dimensions), and \tau, \tau' are exogenous learning rates often set to \tau = 1/\sqrt{2n} and \tau' = 1/\sqrt{2\sqrt{n}} for n-dimensional problems. This derivation stems from the need to maintain positive \sigma_i values while enabling both global (via \tau', promoting uniform scaling) and local (via \tau, allowing dimension-specific tuning) adjustments; the exponential multiplier ensures multiplicative changes that scale with current \sigma_i, fostering progressive refinement from broad to precise as proceeds. Self-adaptation was integrated into EP in the early , drawing parallels to strategies but tailored for EP's -centric framework.

Selection mechanisms

In evolutionary programming (EP), the fitness function provides a problem-specific of an individual's , typically measuring against a defined objective such as prediction accuracy or error minimization. In classical EP applications, such as those involving finite state machines for sequence prediction, is calculated as the fraction of correct predictions over a series of input symbols or as an average payoff derived from a of expected versus actual outputs. These raw values can be used directly or converted to rankings to normalize across the population, ensuring fair assessment regardless of scale. Modern extensions of EP adapt the fitness function to handle complex scenarios, including where trade-offs among conflicting goals (e.g., cost versus performance) are evaluated using Pareto dominance or scalarization techniques. In noisy fitness landscapes, common in real-world simulations or sensor-based problems, evaluations incorporate variations; strategies like averaging multiple fitness assessments or robust mitigate effects to maintain reliable selection. In classical EP, fitness directly represents a payoff, such as the number of correct behavioral predictions in predictive tasks, emphasizing survival based on demonstrated competence. The core selection mechanism in EP is tournament selection, pioneered by David Fogel in 1988, which introduces competition to balance and . Each individual competes against q randomly chosen opponents (often q=10), earning a "win" if its equals or exceeds that of the opponent; the total wins determine survival probability, allowing even suboptimal solutions a chance to persist and avoid premature convergence to local optima. may be optionally applied by guaranteeing the retention of the highest- individual, though the inherent randomness of tournaments frequently preserves top performers without explicit intervention. EP follows a generational replacement model where offspring fully supplant parents, commonly using a (μ, λ) scheme with λ = μ—each of the μ parents generates one mutated offspring, and tournament selection chooses the next μ individuals from the combined pool of 2μ. This comma-style approach, where only compete for advancement, fosters through progressive reduction as lower-fitness variants are culled. Variants like (μ + λ) or (μ, μ) are less prevalent in standard EP, as the pure emphasizes adaptation via competition among new candidates. A simplified pseudocode representation of the tournament selection process in EP is:
# Assume [population](/page/Population) P of size 2μ (μ parents + μ [offspring](/page/Offspring)), each with [fitness](/page/Fitness) f[i]

wins = array of size 2μ, initialized to 0

for i in 0 to 2μ - 1:
    for j in 1 to q:
        k = random index from 0 to 2μ - 1, k ≠ i
        if f[i] >= f[k]:
            wins[i] += 1

# Rank by wins descending (ties broken by [fitness](/page/F Fitness) or randomly)
sort P by wins descending

next_generation = top μ individuals from sorted P
This procedure applies selective pressure proportional to relative performance while introducing variability through random pairings.

Applications

Function and parameter optimization

Evolutionary programming (EP) is widely applied to real-parameter optimization problems, where the goal is to minimize an objective f(\mathbf{x}) with \mathbf{x} \in \mathbb{R}^D over a continuous search space. Standard functions, such as the unimodal f(\mathbf{x}) = \sum_{i=1}^D x_i^2 and the multimodal f(\mathbf{x}) = 10D + \sum_{i=1}^D (x_i^2 - 10\cos(2\pi x_i)), are commonly used to evaluate EP's performance on these tasks. EP demonstrates particular strengths in handling non-differentiable and landscapes, where traditional gradient-based methods fail, by relying on search without requiring information. Its self-adaptation mechanism, which dynamically adjusts step sizes alongside the objective parameters, effectively balances of the search space and of promising regions. The optimization in EP begins with random initialization of a of \mu candidate solutions in \mathbb{R}^D. Each individual generates an offspring through , typically using Gaussian or Cauchy distributions, followed by self-adaptive updates to strategy parameters. A selection then chooses the \mu fittest individuals from the combined parent-offspring pool to form the next generation, iterating until convergence criteria are met. In contexts, EP has been employed for optimization, coupling with models to reduce computational costs. EP variants have shown competitive on suites. EP's is assessed via metrics like speed, measured by the number of evaluations to reach a target error, and solution quality, such as mean best over multiple runs. While effective in low- to medium-dimensional spaces (e.g., D \leq 30), EP exhibits limitations in high dimensions without enhancements, as the curse of dimensionality slows and increases sensitivity to initial conditions.

Engineering and control systems

Evolutionary programming (EP) has found significant applications in engineering and control systems, where it evolves adaptive controllers for complex dynamic environments. In particular, EP has been used to design controllers for robots and autonomous vehicles, focusing on behavioral fitness metrics such as , task completion rates, and robustness to environmental perturbations. J. Fogel's pioneering work in the applied EP to evolve finite state machines (FSMs) for autonomous tasks, demonstrating how populations of candidate controllers could be iteratively refined through and selection to predict and respond to dynamic inputs effectively. Fitness evaluation in these control applications typically relies on simulation-based assessments, measuring against criteria like minimization in trajectory following or collision avoidance success rates. For instance, EP has been employed in for high-precision , where evolved models accurately capture nonlinear dynamics in systems. In , EP originated with Fogel's experiments evolving FSMs to predict time-series signals and classify patterns, laying the groundwork for machine intelligence applications. Early implementations successfully recognized handwritten characters by evolving predictive models that adapted to noisy inputs, outperforming random baselines in prediction accuracy on datasets. Modern uses extend to in signals and in processes, where EP generates classifiers that identify deviations in for in industrial systems. Beyond control and recognition, EP addresses diverse engineering challenges, including traffic routing and , where it optimizes flows to minimize congestion in simulated scenarios. In pharmaceutical molecule design, EP evolves molecular structures to maximize affinity targets, aiding pipelines. Military planning applications in the utilized EP for strategic scenario optimization, evolving decision trees for in wargame simulations. Additionally, EP has been applied to cancer detection through evolved neural classifiers that analyze images, achieving diagnostic accuracies competitive with human experts on clinical datasets. A notable case study involves aerospace applications, where EP integrates with trajectory optimization to evolve flight paths for unmanned vehicles, balancing fuel efficiency and mission constraints in multi-objective formulations. As of 2025, recent applications include hybrid EP approaches combined with deep learning for optimization in renewable energy systems, such as wind turbine control. Despite these successes, EP faces challenges in engineering contexts, particularly scalability to real-time systems, where computational demands of population-based evolution limit deployment on embedded hardware. The reliance on simulation-based fitness introduces the "reality gap," requiring careful bridging techniques to ensure evolved controllers perform reliably in physical environments. Addressing these issues often involves hybrid evaluation strategies to balance exploration and exploitation in high-dimensional search spaces.

Variants and extensions

Classical versus modern EP

Classical evolutionary programming (EP), developed in the 1960s by Lawrence J. Fogel, employed finite state machines (FSMs) as the primary representation for solutions, targeting discrete prediction tasks such as time-series forecasting. Individuals in the population were FSMs that modeled behavioral responses to environmental inputs, with evolution driven by simple operators, such as flipping transitions in the or altering output symbols. Population sizes were typically small, reflecting computational constraints of the era, and the overarching goal was to simulate through evolved predictive behaviors rather than direct optimization. In contrast, modern EP, emerging post-1980s under influences from David B. Fogel and others, shifted to real-valued vector representations for individuals, emphasizing problems like function minimization. Variation relies on self-adaptive Gaussian mutations, where mutation step sizes are encoded in the individual's and co-evolved alongside the objective parameters, enabling dynamic adjustment to the search landscape. Selection typically uses mechanisms, such as pairwise competitions based on , to promote fitter solutions without recombination. Contemporary implementations leverage larger populations and incorporate parallelization to handle high-dimensional problems efficiently. This marks a key shift from symbolic and behavioral —where FSMs directly encoded "programs" for —to numerical search techniques focused on tuning in real-valued spaces, enhancing for complex tasks but diminishing the interpretability of evolved solutions as explicit programs. Modern EP's self-adaptation and selection facilitate broader applicability, though it trades the classical approach's strength in , -based tasks, such as , for superior handling of continuous benchmarks. Performance comparisons highlight these differences: classical EP excels in sequence prediction due to its FSM , achieving reliable learning on symbolic with minimal resources, whereas modern variants demonstrate improved on continuous benchmarks compared to earlier non-adaptive methods. As of 2025, classical EP persists primarily as a historical foundation in literature, while modern EP remains a niche within the broader field, supported by open-source tools like the DEAP library, which enables flexible implementations of self-adaptive and tournament-based EP for research and applications.

Hybrid and integrated approaches

Hybrid approaches in evolutionary programming (EP) frequently incorporate elements from evolution strategies (ES) and genetic algorithms (GA) to overcome EP's limitations in handling large-scale or discrete optimization problems. Cooperative coevolution, where the search space is decomposed into subproblems evolved in parallel populations, has been integrated into hybrid EP-ES frameworks to improve scalability for high-dimensional continuous optimization, as demonstrated in applications to real-world engineering design tasks. Similarly, introducing crossover operators from GA into EP variants enhances solution recombination for discrete problems, such as scheduling or combinatorial tasks, leading to better exploration of rugged fitness landscapes compared to mutation-only EP. Integration of EP with neural networks, often termed , leverages EP's mutation-based evolution to optimize network architectures and weights, particularly effective for tasks where gradient-based methods falter, such as non-differentiable objectives. In the , extensions of EP were developed to evolve and recurrent neural networks for and problems, achieving competitive performance against traditional in sequential decision-making scenarios. Building on this, advancements since the 2020s have applied EP to hyperparameter tuning and in models, where EP evolves parameters like learning rates and network configurations to improve performance in tasks such as image classification, often outperforming methods. Other notable extensions include memetic EP, which embeds local search heuristics—such as hill-climbing or —within the evolutionary loop to exploit promising regions intensively, resulting in faster convergence for multimodal functions without sacrificing global search capabilities. Parallel EP adaptations distribute population evaluations across multiple processors or clusters, enabling efficient handling of computationally intensive simulations in environments and achieving near-linear speedups for population sizes exceeding 10,000 individuals. In multi-objective contexts, EP has been hybridized with dominance-based selection from algorithms like SPEA2, incorporating archive maintenance and fitness assignment to approximate Pareto fronts in problems with conflicting goals, such as in engineering systems. As of 2025, modern trends emphasize EP's role in (RL) for policy evolution, where EP-RL methods evolve parameterized policies alongside gradient updates to enhance exploration in sparse-reward environments, as seen in Evolutionary Policy Optimization (EPO) that combines EA diversity with RL stability for robotic control tasks. Open-source tools like DEAP facilitate EP implementations by supporting customizable operators and integration with libraries, allowing practitioners to evolve models for tasks ranging from to game without extensive coding. These integrations yield benefits such as superior handling of deceptive landscapes and improved in non-stationary problems, though challenges persist in balancing computational overhead and tuning the interplay between global and local components to avoid premature convergence.

References

  1. [1]
    An Overview of Evolutionary Programming | SpringerLink
    Evolutionary programming is a method for simulating evolution that has been investigated for over 35 years.
  2. [2]
    Q1.2: What's Evolutionary Programming (EP)?
    Fogel in 1960, is a stochastic OPTIMIZATION strategy similar to GENETIC ALGORITHMs, but instead places emphasis on the behavioral linkage between PARENTS and ...
  3. [3]
    Evolutionary programming - Scholarpedia
    Apr 10, 2011 · Fogel crafted a series of experiments in which finite state machines (FSMs) represented individual organisms in a population of problem solvers.
  4. [4]
    [PDF] An Evolutionary Programming Approach to Self-Adaptation on Finite ...
    David B. Fogel​​ Abstract. Evolutionary programming was first offered as an alternative method for generating artificial intelligence. Experiments were offered ...Missing: definition | Show results with:definition
  5. [5]
    [PDF] Artificial Intelligence Through Simulated Evolution
    To make things more interesting Fogel et al increased the length of recall and gave a bonus for predicting a rare event. • So the score given for predicting a 1 ...
  6. [6]
    Evolutionary Paradigms | SpringerLink
    The best recognized classes of EC algorithms are described, such as Evolution Strategies, Genetic Algorithms, Genetic Programming, Evolutionary Programming, and ...
  7. [7]
    (PDF) A history of evolutionary computation - ResearchGate
    This section presents a brief but comprehensive summary of the history of evolutionary computation, starting with the ground-breaking work of the 1950s.<|control11|><|separator|>
  8. [8]
    [PDF] An Evolutionary Programming Approach to Self-Adaptation on ...
    An Evolutionary Programming Approach to Self-Adaptation on Finite State Machines · L. Fogel, P. Angeline, D. Fogel · Published in Evolutionary Programming 1995 ...
  9. [9]
    David B. Fogel - Engineering and Technology History Wiki
    Jan 21, 2016 · A pioneer in evolutionary programming, Dr. David B. Fogel has helped develop important theoretical foundations and applications ranging from signal processing.Missing: developments 1980s Joshua
  10. [10]
    (PDF) Parallelism and evolutionary algorithms - ResearchGate
    Aug 9, 2025 · This paper contains a modern vision of the parallelization techniques used for evolutionary algorithms (EAs).
  11. [11]
    Evolutionary learning in neural networks by heterosynaptic plasticity
    May 16, 2025 · We explore evolutionary algorithms (EAs) combined with heterosynaptic plasticity as a gradient-free alternative.
  12. [12]
    [PDF] Modern Evolutionary Programming and its Applications
    Data structures such as real-valued vectors, permutations, variable-length vectors, binary strings, and so on were subjected to the same general process of ...
  13. [13]
    Evolutionary Programming | Clever Algorithms
    Evolutionary Programming was developed by Lawrence Fogel, outlined in early papers ... Fogel, "Autonomous automata", Industrial Research, 1962. [Fogel1964], L. J. ...
  14. [14]
    Evolutionary programming: an introduction and some current ...
    Fogel, L. J., Owens, A. J. and Walsh, M. J. (1966) Artificial Intelligence through Simulated Evolution. John Wiley, New York. Google Scholar. Fogel, L. J. ...<|control11|><|separator|>
  15. [15]
    [PDF] An Overview of Evolutionary Algorithms for Parameter Optimization
    Fogel, D. B. (1992a). An analysis of evolutionary programming. In Fogel and Atmar (1992), (pp. 43–51). Fogel, D. B., & Atmar, J. W. (Eds.). (1992) ...
  16. [16]
  17. [17]
    Multiobjective evolutionary algorithms: A survey of the state of the art
    By evolving a population of solutions, multiobjective evolutionary algorithms (MOEAs) are able to approximate the Pareto optimal set in a single run.
  18. [18]
    Evolutionary algorithms in noisy environments: theoretical issues ...
    This paper is devoted to the effects of fitness noise in evolutionary algorithms (EAs). After a short introduction to the history of this research field, ...
  19. [19]
    Elitism in Evolutionary Algorithms | Baeldung on Computer Science
    Mar 18, 2024 · The main purpose of using elitism in evolutionary algorithms is to keep the reference for promising areas of the search space across the generations.
  20. [20]
    (μ,λ)-Evolution Strategy - Algorithm Afternoon
    The (μ,λ)-Evolution Strategy maintains a population of μ candidate solutions, called parents. In each iteration, the algorithm generates λ new candidate ...
  21. [21]
    Evolutionary Algorithms for Parameter Optimization—Thirty Years ...
    Jun 1, 2023 · We address some major developments in the field of evolutionary algorithms, with applications in parameter optimization, over these 30 years.<|separator|>
  22. [22]
  23. [23]
    [PDF] Artificial Intelligence through Evolutionary Programming - DTIC
    'This report examines the manner by which evolutionary programming treats an arbitrary prediction problem. Additional experiments were conducted to clarify.
  24. [24]
    Modern Evolutionary Programming and its Applications
    Evolutionary programming has been used to solve a wide range of engineering problems, including traffic routing and planning, pharmaceutical design ...
  25. [25]
    Hybrid Evolutionary Algorithm for the Optimization of Interplanetary ...
    Hybrid Evolutionary Algorithm for the Optimization of Interplanetary Trajectories · Towards the Integration of Surrogate Models in the Robust Optimization of ...
  26. [26]
    Fitness functions in evolutionary robotics: A survey and analysis
    This paper surveys fitness functions used in the field of evolutionary robotics (ER). Evolutionary robotics is a field of research that applies artificial ...
  27. [27]
    [PDF] Evolutionary Optimisation of Real-Time Systems and Networks - arXiv
    Different optimisation problems will obviously require different fitness functions, and most will be sensitive to different chromosome formats, population sizes ...Missing: scalability | Show results with:scalability<|control11|><|separator|>
  28. [28]
    [PDF] DEAP: Evolutionary Algorithms Made Easy
    DEAP is a novel evolutionary computation framework for rapid prototyping and testing of ideas. Its design departs from most other existing frameworks in ...
  29. [29]
    Hybrid Evolutionary Algorithms | SpringerLink
    This book deals with the theoretical and methodological aspects, as well as various applications to many real world problems from science, technology, business ...
  30. [30]
    Hybrid Multi-Evolutionary Algorithm to Solve Optimization Problems
    Feb 24, 2020 · In this paper, a new Hybrid Multi-Evolutionary Algorithm, that combines the Genetic Algorithm and Evolutionary Strategy is proposed, so it is ...
  31. [31]
    [PDF] Neuroevolution in Deep Neural Networks - arXiv
    Jun 9, 2020 · Neuroevolution is the automated configuration and training of DNNs using Evolutionary Algorithms (EAs), which are nature-inspired techniques.
  32. [32]
    Optimizing deep learning hyper-parameters through an evolutionary ...
    To address this, Multi-node Evolutionary Neural Networks for Deep Learning (MENNDL) is proposed as a method for automating network selection on computational ...Missing: 2020s | Show results with:2020s
  33. [33]
    Memetic algorithms outperform evolutionary algorithms in ...
    Memetic algorithms integrate local search into an evolutionary algorithm to combine the advantages of rapid exploitation and global optimisation.
  34. [34]
    [2304.05811] A Survey on Distributed Evolutionary Computation
    Apr 12, 2023 · Thanks to the intrinsic parallelism of evolutionary computation (EC), it is natural to implement EC on parallel and distributed computing ...
  35. [35]
    [2503.19037] Evolutionary Policy Optimization - arXiv
    Mar 24, 2025 · We propose Evolutionary Policy Optimization (EPO), a hybrid algorithm that combines the scalability and diversity of EAs with the performance and stability of ...Missing: 2020s | Show results with:2020s
  36. [36]
    PyGAD - Python Genetic Algorithm! — PyGAD 3.5.0 documentation
    PyGAD is an open-source Python library for building the genetic algorithm and optimizing machine learning algorithms. It works with Keras and PyTorch.Pygad Module · Pygad.gann Module · Pygad.torchga Module · Pygad.gacnn ModuleMissing: hybrid | Show results with:hybrid
  37. [37]
    [PDF] 1 Hybrid Evolutionary Algorithms: Methodologies, Architectures, and ...
    In the first phase, a hybrid evo- lutionary algorithm is used to confine the search space by evolving a pool of good candidate rules. Genetic programming [32] ...