Work

Kernels for sequentially ordered data

Year: 2019

Type: article

Abstract: We present a novel framework for learning with sequential data of any kind, such as multivariate time series, strings, or sequences of graphs. The main result is a ”sequentialization” \nthat transforms any kernel on a given domain into a kernel for sequences in that domain. \nThis procedure preserves properties such as positive definiteness, the associated kernel feature map is an ordered variant of sample (cross-)moments, and this sequentialized kernel \nis consistent in the sense that it converges to a kernel for paths if sequences converge to \npaths (by discretization). Further, classical kernels for sequences arise as special cases of \nthis method. We use dynamic programming and low-rank techniques for tensors to provide \nefficient algorithms to compute this sequentialized kernel.

Source: UCL Discovery (University College London)

Authors: Franz J. Király, Harald Oberhauser

Language: English

Topic: Time Series Analysis and Forecasting

Subfield: Signal Processing

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Engineering and Physical Sciences Research Council

Awards: G1633672591, G444630160, The Alan Turing Institute

OpenAlex ID: W2963611658

FWCI: 6.908

Cites:

Cited by: 72

Related to: 20