Learning to Mutate with Hypergradient Guided Population
NIPS 2020, 2020.
We show the hyperparameter and mutation schedule learned by hyperparameter mutation in Fig. 5b and Fig. 5c
Computing the gradient of model hyperparameters, i.e., hypergradient, enables a promising and natural way to solve the hyperparameter optimization task. However, gradient-based methods could lead to suboptimal solutions due to the non-convex nature of optimization in a complex hyperparameter space. In this study, we propose a hyperparamet...More
PPT (Upload PPT)