Signatory: differentiable computations of the signature and logsignature transforms, on both CPU and GPU

Conference item

Signatory is a library for calculating and performing functionality related to the signature and logsignature transforms. The focus is on machine learning, and as such includes features such as CPU parallelism, GPU support, and backpropagation. To our knowledge it is the first GPU-capable library for these operations. Signatory implements new features not available in previous libraries, such as efficient precomputation strategies. Furthermore, several novel algorithmic improvements are introduced, producing substantial real-world speedups even on the CPU without parallelism. The library operates as a Python wrapper around C++, and is compatible with the PyTorch ecosystem. It may be installed directly via \texttt{pip}. Source code, documentation, examples, benchmarks and tests may be found at \texttt{\url{https://github.com/patrick-kidger/signatory}}. The license is Apache-2.0.

Authors: Kidger, P; Lyons, TJ

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: OpenReview
• Host title: Proceedings of the International Conference on Learning Representations (ICLR 2021)
• Article number: Poster 2220
• Publication date: 2021-05-03
• Acceptance date: 2021-01-07
• Event title: International Conference on Learning Representations (ICLR 2021)
• Event location: Online
• Event website: https://openreview.net/group?id=ICLR.cc/2021/Conference
• Event start date: 2021-05-03
• Event end date: 2021-05-07
• Language: English
• Keywords: open source; signature; FFR; library; logsignature; gpu