Weighted signature kernels
Suppose that γ and σ are two continuous bounded variation paths which take values in a finite-dimensional inner product space V. The recent papers respectively introduced the truncated and the untruncated signature kernel of γ and σ, and showed how these concepts can be used in classification and pr...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Institute of Mathematical Statistics
2024
|
| _version_ | 1797112558417608704 |
|---|---|
| author | Cass, T Lyons, T Xu, X |
| author_facet | Cass, T Lyons, T Xu, X |
| author_sort | Cass, T |
| collection | OXFORD |
| description | Suppose that γ and σ are two continuous bounded variation paths which
take values in a finite-dimensional inner product space V. The recent papers respectively introduced the truncated and the untruncated signature kernel
of γ and σ, and showed how these concepts can be used in classification and prediction tasks involving multivariate time series. In this paper, we introduce signature
kernels K
γ,σ
φ
indexed by a weight function φ which generalise the ordinary signature kernel. We show how K
γ,σ
φ
can be interpreted in many examples as an average
of PDE solutions, and thus we show how it can be estimated computationally using
suitable quadrature formulae. We extend this analysis to derive closed-form formulae for expressions involving the expected (Stratonovich) signature of Brownian
motion. In doing so we articulate a novel connection between signature kernels
and the notion of the hyperbolic development of a path, which has been a broadly
useful tool in the recent analysis of the signature. As
applications we evaluate the use of different general signature kernels as a basis for
non-parametric goodness-of-fit tests to Wiener measure on path space.
|
| first_indexed | 2024-03-07T08:25:51Z |
| format | Journal article |
| id | oxford-uuid:33382721-40b3-446a-88d7-8b4143737b3f |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T08:25:51Z |
| publishDate | 2024 |
| publisher | Institute of Mathematical Statistics |
| record_format | dspace |
| spelling | oxford-uuid:33382721-40b3-446a-88d7-8b4143737b3f2024-02-20T13:34:45ZWeighted signature kernelsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:33382721-40b3-446a-88d7-8b4143737b3fEnglishSymplectic ElementsInstitute of Mathematical Statistics2024Cass, TLyons, TXu, XSuppose that γ and σ are two continuous bounded variation paths which take values in a finite-dimensional inner product space V. The recent papers respectively introduced the truncated and the untruncated signature kernel of γ and σ, and showed how these concepts can be used in classification and prediction tasks involving multivariate time series. In this paper, we introduce signature kernels K γ,σ φ indexed by a weight function φ which generalise the ordinary signature kernel. We show how K γ,σ φ can be interpreted in many examples as an average of PDE solutions, and thus we show how it can be estimated computationally using suitable quadrature formulae. We extend this analysis to derive closed-form formulae for expressions involving the expected (Stratonovich) signature of Brownian motion. In doing so we articulate a novel connection between signature kernels and the notion of the hyperbolic development of a path, which has been a broadly useful tool in the recent analysis of the signature. As applications we evaluate the use of different general signature kernels as a basis for non-parametric goodness-of-fit tests to Wiener measure on path space. |
| spellingShingle | Cass, T Lyons, T Xu, X Weighted signature kernels |
| title | Weighted signature kernels |
| title_full | Weighted signature kernels |
| title_fullStr | Weighted signature kernels |
| title_full_unstemmed | Weighted signature kernels |
| title_short | Weighted signature kernels |
| title_sort | weighted signature kernels |
| work_keys_str_mv | AT casst weightedsignaturekernels AT lyonst weightedsignaturekernels AT xux weightedsignaturekernels |