Score: 2

A Common Interface for Automatic Differentiation

Published: May 8, 2025 | arXiv ID: 2505.05542v1

By: Guillaume Dalle, Adrian Hill

Potential Business Impact:

Lets scientists easily test different math tools.

Business Areas:
Machine Learning Artificial Intelligence, Data and Analytics, Software

For scientific machine learning tasks with a lot of custom code, picking the right Automatic Differentiation (AD) system matters. Our Julia package DifferentiationInterface$.$jl provides a common frontend to a dozen AD backends, unlocking easy comparison and modular development. In particular, its built-in preparation mechanism leverages the strengths of each backend by amortizing one-time computations. This is key to enabling sophisticated features like sparsity handling without putting additional burdens on the user.

Country of Origin
🇩🇪 🇫🇷 Germany, France


Page Count
11 pages

Category
Computer Science:
Mathematical Software