arXiv · 2311.18727
Automatic Functional Differentiation in JAX
Abstract
We extend JAX with the capability to automatically differentiate higher-order functions (functionals and operators). By representing functions as a generalization of arrays, we seamlessly use JAX's existing primitive system to implement higher-order functions. We present a set of primitive operators that serve as foundational building blocks for constructing several key types of functionals. For every introduced primitive operator, we derive and implement both linearization and transposition rules, aligning with JAX's internal protocols for forward and reverse mode automatic differentiation. This enhancement allows for functional differentiation in the same syntax traditionally use for functions. The resulting functional gradients are themselves functions ready to be invoked in python. We showcase this tool's efficacy and simplicity through applications where functional derivatives are indispensable. The source code of this work is released at https://github.com/sail-sg/autofd .
Explore related subjects
Keep this discovery
Min Lin. 2023-11-30. Automatic Functional Differentiation in JAX. https://arxiv.org/abs/2311.18727
Cite the original work for its findings. Save a collection to share your selection of sources.