Reliable and Fast Recurrent Neural Network Architecture Optimization

Abstract

This article introduces Random Error Sampling-based Neuroevolution (RESN), a novel automatic method to optimize recurrent neural network architectures. RESN combines an evolutionary algorithm with a training-free evaluation approach. The results show that RESN achieves state-of-the-art error performance while reducing by half the computational time.

References in corpus (2)

• On the difficulty of training Recurrent Neural Networks
• Metaheuristic Design of Feedforward Neural Networks: A Review of Two Decades of Research