Can I read Zero Grads: Learning Local Surrogate Losses for Non-Differentiable Graphics on EtoBox?

Zero Grads: Learning Local Surrogate Losses for Non-Differentiable Graphics by Fischer, Michael; Ritschel, Tobias is a scholarly article available to read on EtoBox.

What is Zero Grads: Learning Local Surrogate Losses for Non-Differentiable Graphics about?

Gradient-based optimization is now ubiquitous across graphics, but unfortunately can not be applied to problems with undefined or zero gradients. To circumvent this issue, the loss function can be manually replaced by a ``surrogate'' that has similar minima but is differentiable. Our proposed framework, ZeroGrads, automates this process by learning a neural approximation of the objective function, which in turn can be used to differentiate through arbitrary black-box graphics pipelines. We train the surrogate on an actively smoothed version of the objective and encourage locality, focusing the surrogate's capacity on what matters at the current training episode. The fitting is performed online, alongside the parameter optimization, and self-supervised, without pre-computed data or pre-trained models. As sampling the objective is expensive (it requires a full rendering or simulator run), we devise an efficient sampling scheme that allows for tractable run-times and competitive performance at little overhead. We demonstrate optimizing diverse non-convex, non-differentiable black-box problems in graphics, such as visibility in rendering, discrete parameter spaces in procedural modelli

Author

Fischer, Michael; Ritschel, Tobias

Published

2023

Language

EN