Evolutionary acquisition of neural topologies¶

An evolutionary reinforcement-learning method that jointly evolves artificial-neural-network structure and weights, using structural mutation and evolution-strategy parameter optimization.

Version: v2 · 2026-09-19 · History
Domain-specific #: 4447
Domain group: Applied Sciences & Engineering
Origin domain: Computer Science & Software Engineering
Subdomains: Machine Learning, Neuroevolution → Computer Science & Software Engineering

Core Idea¶

Evolutionary acquisition of neural topologies is a neuroevolution method in which both network connections and their numerical parameters are optimized by evolutionary search.^[1] Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

The load-bearing residual is not the broad topic of machine learning. It is EANT-family coupling of topology growth with evolution-strategy weight adaptation. That residual remains recognizable when examples, notation, scale, or implementation change, but it disappears if the carrier is mistyped, the condition that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators fails, a neighboring object is substituted, or notation and topical resemblance replace the constitutive test. This gives the entry an operational identity rather than merely a historical label.

A useful analysis keeps three layers separate. The constitutive layer says what must be true: topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators. The evidential layer asks what observation or proof warrants the claim: type the carrier, state every parameter and convention in the definition, test that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases. The use layer asks what reasoning becomes available once the identity is established: recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions. Conflating the layers is the most common source of scope inflation.

How would you explain it like I'm…

Growing Bead Brains

Imagine building little thinking machines out of beads and strings. You start with tiny ones, then make copies with an extra string added or a string tugged tighter. You keep the ones that do the job best and try again, so both the shape and the tightness get better together.

Evolving Network Shapes

Evolutionary acquisition of neural topologies is a way for computers to design brain-like networks by copying how evolution works. It starts with small, simple networks. New versions are made by changes that add or alter connections, and other changes that adjust the numbers, or weights, on those connections. The networks that do the task best are kept to make the next round. The network's layout and its numbers are passed down together.

Coupled Topology-and-Weight Evolution

Evolutionary acquisition of neural topologies (EANT) is a neuroevolution method: it uses evolutionary search to optimize both a neural network's structure (its topology, which nodes connect) and its numerical weights. Networks start compact; mutations add or modify structure, while evolution-strategy operators fine-tune the weights. Selection keeps the networks with higher fitness on the task. A key feature is that topology and weights are inherited together as one genotype, rather than training weights separately with something like backpropagation.

Evolutionary acquisition of neural topologies is a neuroevolution method that jointly optimizes network connectivity and connection parameters by evolutionary search. Individuals begin with minimal structures and grow complexity through structural mutations that add or modify topology. Weights are adapted by evolution-strategy operators rather than by gradient-based training. Selection retains individuals with higher fitness measured in a declared task environment. What identifies the EANT family is this coupling of topology growth with evolution-strategy weight adaptation, with topology and weights kept within a single heritable genotype. If structure and weights are optimized by decoupled, non-heritable means, or fitness is not evaluated through the declared environment and operators, the method is something else, even if superficially similar.

Structural Signature¶

Carrier: a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections
Inputs or antecedent state: the exact machine learning carrier, defining parameters and conventions, boundary conditions, source evidence, comparison cases, and any measurement or proof assumptions needed to evaluate Evolutionary acquisition of neural topologies
Constitutive operation: Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness.
Invariant: topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators
Recognition test: type the carrier, state every parameter and convention in the definition, test that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases
Output or consequence: recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions
Failure boundary: the carrier is mistyped, the condition that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators fails, a neighboring object is substituted, or notation and topical resemblance replace the constitutive test

What It Is Not¶

It is not the whole field of machine learning. The field contains many questions and methods that do not instantiate Evolutionary acquisition of neural topologies.
It is not its most familiar example. A controller population evolves new recurrent links while CMA-ES-like steps refine the weights of each candidate network. exhibits the structure, but the example is evidence for the abstraction rather than its definition.
It is not the neighboring catalog concept NeuroEvolution of Augmenting Topologies. Both evolve topology and weights, but EANT uses its own structural representation and evolution-strategy parameter mutation rather than NEAT's historical markings and speciation design.
It is not a claim that every boundary case has one uncontested classification. a generalized or degenerate case may change existence, uniqueness, measurement, or naming conventions, so the exact definition of Evolutionary acquisition of neural topologies must control the decision
It is not an unrestricted metaphor for any process that seems similar. Outside machine learning, the vocabulary and validity conditions do not transfer literally.

Scope of Application¶

Evolutionary acquisition of neural topologies belongs to machine learning and is useful where the analyst can specify a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections, then evaluate topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators. The scope is broad within that domain but bounded by the need for topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators. The entry records a descriptive analytical identity; practical use requires the governing domain's evidence, standards, and safety obligations.^[2]

Definition and recognition. Determine whether a proposed instance satisfies the constitutive conditions rather than merely sharing terminology.
Construction or evolution. Track how the exact machine learning carrier, defining parameters and conventions, boundary conditions, source evidence, comparison cases, and any measurement or proof assumptions needed to evaluate Evolutionary acquisition of neural topologies are converted, constrained, or organized by Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness..
Comparison. Compare instances using carrier, parameters, convention, domain, scale, boundary conditions, evidence, exact versus approximate form, and limiting behavior, without treating convenience measures as the definition.
Boundary analysis. Diagnose cases where a generalized or degenerate case may change existence, uniqueness, measurement, or naming conventions, so the exact definition of Evolutionary acquisition of neural topologies must control the decision and state which convention or theorem controls the decision.
Downstream reasoning. Use the established identity to support recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions while preserving the assumptions under which the inference is valid.

Clarity¶

The abstraction clarifies a crowded vocabulary by making topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Evolutionary acquisition of neural topologies can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated. The disciplined statement is: given the exact machine learning carrier, defining parameters and conventions, boundary conditions, source evidence, comparison cases, and any measurement or proof assumptions needed to evaluate Evolutionary acquisition of neural topologies, the structure counts as Evolutionary acquisition of neural topologies exactly when topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators.

This format also separates identity from measurement. Empirical, computational, or documentary proxies support recognition only under declared validity and uncertainty assumptions; formal cases require proof rather than measurement. Measurements can be noisy, implementations can approximate, and proofs can use equivalent characterizations; none of those facts licenses changing the object being measured. When reports disagree, first check scope and convention, then data or proof, and only then interpret the disagreement as substantive.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Evolutionary acquisition of neural topologies. Evolutionary acquisition of neural topologies compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

The compression has a price. A single label can hide canonical, generalized, restricted, approximate, computational, empirical, and historically variant formulations of Evolutionary acquisition of neural topologies. Good use therefore carries a small declaration of assumptions alongside the name. The abstraction manages complexity when it reduces the state space of the question while keeping the failure boundary visible; it mismanages complexity when the label substitutes for that boundary analysis.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators independently of one notation or implementation. This step prevents the canonical example from becoming the definition.
Derive consequences. From topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators, infer recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions. Record each assumption used so that a later change of setting does not silently preserve an invalid conclusion.
Test adversarial cases. Examine a generalized or degenerate case may change existence, uniqueness, measurement, or naming conventions, so the exact definition of Evolutionary acquisition of neural topologies must control the decision and an object that resembles Evolutionary acquisition of neural topologies in purpose or vocabulary but does not satisfy its invariant is outside the class. A robust identity explains why the first is convention-sensitive and why the second is outside the class.
Compare and refine. Use carrier, parameters, convention, domain, scale, boundary conditions, evidence, exact versus approximate form, and limiting behavior to compare legitimate instances, and refine the model when discrepancies reflect hidden variation rather than failure of the abstraction itself.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of machine learning because they reuse a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections, Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness., and type the carrier, state every parameter and convention in the definition, test that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases. A theorem, diagnostic, or modeling warning can travel when those roles remain literal. For example, the distinction between constitutive identity and a convenient observable transfers from A controller population evolves new recurrent links while CMA-ES-like steps refine the weights of each candidate network. to Evaluation reports random seeds, population budget, fitness shaping and comparisons with NEAT or fixed-topology baselines..^[3]

Transfer outside the home domain is weaker. The skeletal pattern—type the carrier, apply the defining mechanism of Evolutionary acquisition of neural topologies, preserve its invariant, and derive only consequences licensed by the stated boundary—may suggest an analogy, but the domain-specific mechanisms, admissible evidence, and consequences do not come along automatically. The safe transfer procedure maps each role explicitly, checks the invariant again, and refuses the name when only a superficial resemblance remains.

Examples¶

Canonical¶

A controller population evolves new recurrent links while CMA-ES-like steps refine the weights of each candidate network. The example exposes the carrier and directly tests that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators; changing incidental notation preserves the identity, while removing that condition destroys it. This example is canonical because every role can be inspected: the carrier is a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections; the operative rule is Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness.; the invariant is topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators; and the result supports recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions.^[1] Changing incidental notation or scale leaves the structure intact, while removing topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators destroys the classification.

Mapped back: a population of neural networks, graph topologies and weights, fitness from an environment, mutation operators, adaptive step sizes or CMA-ES, selection, generations and recurrent connections → Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness. → topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators → recognizing and comparing instances of Evolutionary acquisition of neural topologies, deriving its domain-specific consequences, selecting valid models or methods, and preventing transfer beyond its assumptions

Applied / In Practice¶

Evaluation reports random seeds, population budget, fitness shaping and comparisons with NEAT or fixed-topology baselines. The applied case qualifies only because the same invariant and boundary test remain literal under changed parameters or implementation. The applied case is not licensed merely by vocabulary. It qualifies because the same recognition test—type the carrier, state every parameter and convention in the definition, test that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases—can be run and because the same failure boundary—the carrier is mistyped, the condition that topology and weights remain part of one heritable genotype and fitness evaluation uses the declared environment and evolutionary operators fails, a neighboring object is substituted, or notation and topical resemblance replace the constitutive test—remains meaningful.^[2] The case also shows why practical outputs should report assumptions, resolution, and uncertainty instead of a naked label.

Mapped back: declared instance → recognition test → boundary check → qualified use

Structural Tensions¶

T1: Axiomatic identity vs. operational recognition. The defining conditions may be exact while empirical or computational recognition is approximate. Neither pole can be removed without changing the analytical task. Diagnostic: Can the reviewer state both the exact condition and the evidence used to infer it?
T2: Local roles vs. global consequence. The mechanism is enacted through local relations, but the abstraction is usually valued for a global classification or prediction. Neither pole can be removed without changing the analytical task. Diagnostic: Does the claimed global result actually follow from the declared local conditions?
T3: Ideal form vs. finite representation. Theory states a clean invariant while data structures, measurements, or proofs expose only finite representations. Neither pole can be removed without changing the analytical task. Diagnostic: Would increasing resolution converge toward the same classification?
T4: Canonical convention vs. legitimate variants. A standard formulation supports communication, while variants may preserve the same core under changed assumptions. Neither pole can be removed without changing the analytical task. Diagnostic: Which role is invariant across variants, and which convention-specific conclusion changes?
T5: Compression vs. hidden assumptions. The name compresses a complex argument but can conceal prerequisites. Neither pole can be removed without changing the analytical task. Diagnostic: Can each downstream inference be traced to an explicit assumption?
T6: Autonomous residual vs. reduction to catalog neighbors. The candidate uses broader structures but adds an identity-bearing residual. Neither pole can be removed without changing the analytical task. Diagnostic: After subtracting the proposed parent and named neighbors, does the constitutive residual still support independent diagnostics?

Structural–Framed Character¶

The entry is structurally mixed but domain-framed. Its portable skeleton is type the carrier, apply the defining mechanism of Evolutionary acquisition of neural topologies, preserve its invariant, and derive only consequences licensed by the stated boundary. Its identity-bearing terms—Evolutionary acquisition of neural topologies, carrier, parameter, invariant, boundary, evidence, model, transformation, and application—derive their meaning from machine learning and cannot be replaced by generic systems language without losing the tests that distinguish valid from invalid instances.

This mixed character explains why the abstraction is reusable inside the domain yet does not meet the Prime bar. The structure organizes reasoning, but its claims still depend on domain-specific objects, evidence, and intervention semantics.

Structural Core vs. Domain Accent¶

The structural core consists of a carrier, Networks begin with compact structures, mutations add or modify topology, evolution-strategy operators tune weights, and selection retains individuals with higher task fitness., a recognition invariant, and a consequence. That skeleton may resemble patterns elsewhere, especially type the carrier, apply the defining mechanism of Evolutionary acquisition of neural topologies, preserve its invariant, and derive only consequences licensed by the stated boundary. The domain accent is not decorative: Evolutionary acquisition of neural topologies, carrier, parameter, invariant, boundary, evidence, model, transformation, and application determine what counts as an admissible carrier, a valid transition, and successful evidence.

The abstraction therefore remains domain-specific. A cross-domain reuse that preserves only words such as 'balance,' 'cut,' 'sequence,' 'loss,' or 'simulation' is metaphor. Literal transfer requires the original role structure and diagnostics, which in this case remain anchored in machine learning.

This entry is a kind of Optimization.

The method optimizes neural structure and parameters through evolution. What makes it its own entry is the EANT-family coupling of neural-network topology growth with evolution-strategy weight adaptation in one reinforcement-learning method.

Relationships to Other Abstractions¶

Current abstraction Evolutionary acquisition of neural topologies Domain-specific

Parents (1) — outgoing typed relations

Evolutionary acquisition of neural topologies is a kind of Optimization Prime

The method optimizes neural structure and parameters through evolution.

Hierarchy path (1) — routes to 1 parentless root

Evolutionary acquisition of neural topologies → is a kind of Optimization

Neighborhood in Abstraction Space¶

Evolutionary acquisition of neural topologies sits in a crowded region of the domain-specific corpus (25^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Network Models & Graph Generation (6 abstractions)

Nearest neighbors

Evolutionary data mining — 0.91
Premature convergence — 0.90
Fitness model (network theory) — 0.89
Graph neural network — 0.89
Population-based incremental learning — 0.89

Computed from structural-signature embeddings · 2026-10-25

Not to Be Confused With¶

NeuroEvolution of Augmenting Topologies. Both evolve topology and weights, but EANT uses its own structural representation and evolution-strategy parameter mutation rather than NEAT's historical markings and speciation design.
One canonical example. An instance demonstrates the structure but does not define the whole abstraction.
Measurement or implementation of Evolutionary acquisition of neural topologies. A proxy or realization is evidence for the abstraction, not the abstraction itself.
Generalized Evolutionary acquisition of neural topologies. An extension qualifies only when its changed axioms and retained invariant are stated.

References¶

[1] Source cited in the frozen article, 'Archived copy'. registry ↩a ↩b

[2] Peter J Angeline, Gregory M Saunders, and Jordan B Pollack. An evolutionary algorithm that constructs recurrent neural networks. IEEE Transactions on Neural Networks, 5:54–65, 1994. [http://demo.cs.brandeis.edu/papers/ieeenn.pdf]. registry ↩a ↩b

[3] NeuroEvolution of Augmented Topologies (NEAT) by Stanley and Miikkulainen, 2005 [http://nn.cs.utexas.edu/downloads/papers/stanley.ieeetec05.pdf]. registry ↩