Young Researcher and Innovator Conference Grant
NEVO-GSPT: Population-Based Neural Network Evolution Using Inflate and Deflate Operators
Davide Farinati, Universidade Nova de Lisboa, Lisbon, Portugal, and Ospedale San Raffaele, Milan, Italy
Frederico J. J. B. Santos, University of Trieste, Italy
Leonardo Vanneschi, Universidade Nova de Lisboa, Lisbon, Portugal
Mauro Castelli, Universidade Nova de Lisboa, Lisbon, Portugal
29th European Conference on Genetic Programming (EuroGP, part of Evostar 2026)
Toulouse, France
8-10 April 2026
Grantee: Davide Farinati, Universidade Nova de Lisboa, Lisbon, Portugal, and Ospedale San Raffaele, Milan, Italy
Type of publication: Full paper
Type of presentation: Oral and poster
Start date: 2026-04-08
End date: 2026-04-10
Awarded: 2026-02-19
Report approved: 2026-05-12
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 NeuroEVOlution through Geometric Semantic perturbation and Population based Training (NEVO-GSPT), 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, NEVO-GSPT’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 NEVO-GSPT 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.
