https://enzyme.mit.edu/

[logo] Enzyme AD

Enzyme Automatic Differentiation Framework

  * Community
      + Discussion List
  * Source
  * Bugs
  * FAQ

Enzyme Overview

The Enzyme project is a tool for performing reverse-mode automatic
differentiation (AD) of statically-analyzable LLVM IR. This allows
developers to use Enzyme to automatically create gradients of their
source code without much additional work.

double foo(double);

double grad_foo(double x) {
    return __enzyme_autodiff(foo, x);
}

By differentiating code after optimization, Enzyme is able to create
substantially faster derivatives than existing tools that
differentiate programs before optimization.

[all_top]

Components

Enzyme is composed of four pieces:

  * An optional preprocessing phase which performs minor
    transformations that tend to be helpful for AD.
  * A new interprocedural type analysis that deduces the underlying
    types of memory locations
  * An activity analaysis that determines what instructions or values
    can impact the derivative computation (common in existing AD
    systems).
  * An optimization pass which creates any required derivative
    functions, replacing calls to __enzyme_autodiff with the
    generated functions.

More resources

For more information on Enzyme, please see:

  * The Enzyme getting started guide
  * The Enzyme mailing list for any questions.
  * Previous talks .

Citing Enzyme

To cite Enzyme, please cite the following:

@incollection{enzymeNeurips,
title = {Instead of Rewriting Foreign Code for Machine Learning, Automatically Synthesize Fast Gradients},
author = {Moses, William S. and Churavy, Valentin},
booktitle = {Advances in Neural Information Processing Systems 33},
year = {2020},
}

The original Enzyme is also avaiable as a preprint on arXiv .

Next - Installation


Powered by Hugo. Theme by TechDoc. Designed by Thingsym.

  * Home
  * Installation
  * Charter
  * Getting Started+
      + CUDA Guide
      + Using Enzyme
      + Calling Convention
      + FAQ
  * Talks and Related Publications