Kernels for sequentially ordered data

Abstract

We present a novel framework for kernel learning with sequential data of any kind, such as time series, sequences of graphs, or strings. Our approach is based on signature features which can be seen as an ordered variant of sample (cross-)moments; it allows to obtain a "sequentialized" version of any static kernel. The sequential kernels are efficiently computable for discrete sequences and are shown to approximate a continuous moment form in a sampling sense. A number of known kernels for sequences arise as "sequentializations" of suitable static kernels: string kernels may be obtained as a special case, and alignment kernels are closely related up to a modification that resolves their open non-definiteness issue. Our experiments indicate that our signature-based sequential kernel framework may be a promising approach to learning with sequential data, such as time series, that allows to avoid extensive manual pre-processing.

References in corpus (5)

• Sparse arrays of signatures for online character recognition
• Learning from the past, predicting the statistics for the future, learning an evolving system
• Extracting information from the signature of a financial data stream
• Rotation invariants of two dimensional curves based on iterated integrals
• Character-level Chinese Writer Identification using Path Signature Feature, DropStroke and Deep CNN

Cited by in corpus (4)

• Persistence paths and signature features in topological data analysis
• Application of the Signature Method to Pattern Recognition in the CEQUEL Clinical Trial
• Logsig-RNN: a novel network for robust and efficient skeleton-based action recognition
• SigFormer: Signature Transformers for Deep Hedging