Deep signature transforms
File(s)
Published version
Author(s)
Bonnier, Patric
Kidger, Patrick
Perez Arribas, Imanol
Salvi, Cristopher
Lyons, Terry
Type
Conference Paper
Abstract
The signature is an infinite graded sequence of statistics known to characterise a stream of data up to a negligible equivalence class. It is a transform which has previously been treated as a fixed feature transformation, on top of which a model may be built. We propose a novel approach which combines the advantages of the signature transform with modern deep learning frameworks. By learning an augmentation of the stream prior to the signature transform, the terms of the signature may be selected in a data-dependent way. More generally, we describe how the signature transform may be used as a layer anywhere within a neural network. In this context it may be interpreted as a pooling operation. We present the results of empirical experiments to back up the theoretical justification. Code available at \texttt{github.com/patrick-kidger/Deep-Signature-Transforms}.
Date Issued
2019-12-08
Date Acceptance
2019-09-05
Citation
NeurIPS Proceedings, 2019, 32, pp.1-17
ISBN
9781713807933
Start Page
1
End Page
17
Journal / Book Title
NeurIPS Proceedings
Volume
32
Copyright Statement
© 2019 The Author(s).
Identifier
https://papers.nips.cc/paper/2019/hash/d2cdf047a6674cef251d56544a3cf029-Abstract.html
Source
Advances in Neural Information Processing Systems 32 (NeurIPS 2019)
Publication Status
Published
Start Date
2019-12-08
Finish Date
2019-12-14
Coverage Spatial
Vancouver, Canada