Ryan P. Adams

Gradient-based Hyperparameter Optimization through Reversible Learning

2015 · arXiv (Cornell University)

Tuning hyperparameters of learning algorithms is hard because gradients are usually unavailable. We compute exact gradients of cross-validation performance with respect to all hyperparameters by chaining derivatives backwards through the entire training procedure. These gradients …

Representing and Learning Functions Invariant Under Crystallographic Groups

2023 · arXiv (Cornell University)

Crystallographic groups describe the symmetries of crystals and other repetitive structures encountered in nature and the sciences. These groups include the wallpaper and space groups. We derive linear and nonlinear representations of functions that are …