𝜆ₛ: computable semantics for differentiable programming with higher-order functions and datatypes
Abstract
Deep learning is moving towards increasingly sophisticated optimization objectives that employ higher-order functions, such as integration, continuous optimization, and root-finding. Since differentiable programming frameworks such as PyTorch and TensorFlow do not have first-class representations of these functions, developers must reason about the semantics of such objectives and manually translate them to differentiable code.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Jan 04, 2021
- Source ID
- 10.1145/3434284
Entities
People
- Benjamin Sherman
- Jesse Michel
- Michael Carbin
Organizations
- Massachusetts Institute of Technology
- Office of Naval Research