Better call Surrogates: A hybrid Evolutionary Algorithm for Hyperparameter optimization
This work addresses the problem of efficient hyperparameter optimization for machine learning practitioners, offering an incremental improvement over existing evolutionary algorithms.
This paper proposes STEADE, a surrogate-assisted evolutionary algorithm for hyperparameter optimization of machine learning models. STEADE estimates the objective function landscape using Radial Basis Function interpolation and guides a Differential Evolution algorithm with a Bayesian optimization framework, demonstrating improvement over a vanilla EA.
In this paper, we propose a surrogate-assisted evolutionary algorithm (EA) for hyperparameter optimization of machine learning (ML) models. The proposed STEADE model initially estimates the objective function landscape using RadialBasis Function interpolation, and then transfers the knowledge to an EA technique called Differential Evolution that is used to evolve new solutions guided by a Bayesian optimization framework. We empirically evaluate our model on the hyperparameter optimization problems as a part of the black box optimization challenge at NeurIPS 2020 and demonstrate the improvement brought about by STEADE over the vanilla EA.