NEVO-GSPT: Population-Based Neural Network Evolution Using Inflate and Deflate Operators

2026-01-13

AI Summary

This work addresses the challenges of high computational cost and uncontrolled semantic changes during neural architecture evolution by introducing geometric semantic operators (GSOs) into the evolutionary process. It proposes Inflate/Deflate operations to construct an efficient evolutionary framework capable of effectively exploring the architecture space. A newly designed DGSM operator enables dynamic network size reduction while preserving semantic equivalence, significantly accelerating population training by computing semantics only for newly added components. Evaluated on four regression benchmarks, the method evolves compact yet high-performing networks that match or surpass state-of-the-art approaches, including standard neural networks, SLIM-GSGP, TensorNEAT, and SLM.

Technology Category

• Search and Optimization: Evolutionary Computation
• Machine Learning: Evolutionary Learning
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Evolving neural network architectures is a computationally demanding process. Traditional methods often require an extensive search through large architectural spaces and offer limited understanding of how structural modifications influence model behavior. This paper introduces \gls{ngspt}, a novel Neuroevolution algorithm based on two key innovations. First, we adapt geometric semantic operators~(GSOs) from genetic programming to neural network evolution, ensuring that architectural changes produce predictable effects on network semantics within a unimodal error surface. Second, we introduce a novel operator (DGSM) that enables controlled reduction of network size, while maintaining the semantic properties of~GSOs. Unlike traditional approaches, \gls{ngspt}'s efficient evaluation mechanism, which only requires computing the semantics of newly added components, allows for efficient population-based training, resulting in a comprehensive exploration of the search space at a fraction of the computational cost. Experimental results on four regression benchmarks show that \gls{ngspt} consistently evolves compact neural networks that achieve performance comparable to or better than established methods in the literature, such as standard neural networks, SLIM-GSGP, TensorNEAT, and SLM.

Problem

• neural architecture evolution
• computational efficiency
• structural modification
• search space exploration
• neuroevolution

Innovation

• Neuroevolution
• Geometric Semantic Operators
• Network Pruning
• Population-Based Evolution
• Efficient Evaluation