Dynamics-adapted cone kernels

Research output: Contribution to journalArticle

Abstract

We present a family of kernels for analysis of data generated by dynamical systems these so-called cone kernels feature a dependence on the dynamical vector field operating in the phase space manifold, estimated empirically through finite differences of time-ordered data samples. In particular, cone kernels assign strong affinity to pairs of samples whose relative displacement vector lies within a narrow cone aligned with the dynamical vector field the outcome of this explicit dependence on the dynamics is that, in a suitable asymptotic limit, Laplace-Beltrami operators for data analysis constructed from cone kernels generate diffusions along the integral curves of the dynamical vector field. This property is independent of the observation modality, and endows these operators with invariance under a weakly restrictive class of transformations of the data (including conformal transformations), while it also enhances their capability to extract intrinsic dynamical timescales via eigenfunctions. Here, we study these features by establishing the Riemannian metric tensor induced by cone kernels in the limit of large data. We find that the corresponding Dirichlet energy is governed by the directional derivative of functions along the dynamical vector field, giving rise to a measure of roughness of functions that favors slowly varying observables. We demonstrate the utility of cone kernels in nonlinear flows on the 2-torus and North Pacific sea surface temperature data generated by a comprehensive climate model.

Original languageEnglish (US)
Pages (from-to)556-608
Number of pages53
JournalSIAM Journal on Applied Dynamical Systems
Volume14
Issue number2
DOIs
StatePublished - 2015

Fingerprint

Cones
Cone
kernel
Vector Field
Climate models
Conformal Transformation
Data Transformation
Sea Surface Temperature
Climate Models
Laplace-Beltrami Operator
Directional derivative
Asymptotic Limit
Large Data
Riemannian Metric
Invariance
Eigenvalues and eigenfunctions
Roughness
Modality
Dirichlet
Affine transformation

Keywords

  • Delay Coordinates
  • Diffusion Operators
  • Eigenfunctions
  • Kernel Methods
  • Manifold Embedding
  • Vector Field

ASJC Scopus subject areas

  • Analysis
  • Modeling and Simulation

Cite this

Dynamics-adapted cone kernels. / Giannakis, Dimitrios.

In: SIAM Journal on Applied Dynamical Systems, Vol. 14, No. 2, 2015, p. 556-608.

Research output: Contribution to journalArticle

@article{758e034f352b4da887787e0c43cbe6e6,
title = "Dynamics-adapted cone kernels",
abstract = "We present a family of kernels for analysis of data generated by dynamical systems these so-called cone kernels feature a dependence on the dynamical vector field operating in the phase space manifold, estimated empirically through finite differences of time-ordered data samples. In particular, cone kernels assign strong affinity to pairs of samples whose relative displacement vector lies within a narrow cone aligned with the dynamical vector field the outcome of this explicit dependence on the dynamics is that, in a suitable asymptotic limit, Laplace-Beltrami operators for data analysis constructed from cone kernels generate diffusions along the integral curves of the dynamical vector field. This property is independent of the observation modality, and endows these operators with invariance under a weakly restrictive class of transformations of the data (including conformal transformations), while it also enhances their capability to extract intrinsic dynamical timescales via eigenfunctions. Here, we study these features by establishing the Riemannian metric tensor induced by cone kernels in the limit of large data. We find that the corresponding Dirichlet energy is governed by the directional derivative of functions along the dynamical vector field, giving rise to a measure of roughness of functions that favors slowly varying observables. We demonstrate the utility of cone kernels in nonlinear flows on the 2-torus and North Pacific sea surface temperature data generated by a comprehensive climate model.",
keywords = "Delay Coordinates, Diffusion Operators, Eigenfunctions, Kernel Methods, Manifold Embedding, Vector Field",
author = "Dimitrios Giannakis",
year = "2015",
doi = "10.1137/140954544",
language = "English (US)",
volume = "14",
pages = "556--608",
journal = "SIAM Journal on Applied Dynamical Systems",
issn = "1536-0040",
publisher = "Society of Industrial and Applied Mathematics",
number = "2",

}

TY - JOUR

T1 - Dynamics-adapted cone kernels

AU - Giannakis, Dimitrios

PY - 2015

Y1 - 2015

N2 - We present a family of kernels for analysis of data generated by dynamical systems these so-called cone kernels feature a dependence on the dynamical vector field operating in the phase space manifold, estimated empirically through finite differences of time-ordered data samples. In particular, cone kernels assign strong affinity to pairs of samples whose relative displacement vector lies within a narrow cone aligned with the dynamical vector field the outcome of this explicit dependence on the dynamics is that, in a suitable asymptotic limit, Laplace-Beltrami operators for data analysis constructed from cone kernels generate diffusions along the integral curves of the dynamical vector field. This property is independent of the observation modality, and endows these operators with invariance under a weakly restrictive class of transformations of the data (including conformal transformations), while it also enhances their capability to extract intrinsic dynamical timescales via eigenfunctions. Here, we study these features by establishing the Riemannian metric tensor induced by cone kernels in the limit of large data. We find that the corresponding Dirichlet energy is governed by the directional derivative of functions along the dynamical vector field, giving rise to a measure of roughness of functions that favors slowly varying observables. We demonstrate the utility of cone kernels in nonlinear flows on the 2-torus and North Pacific sea surface temperature data generated by a comprehensive climate model.

AB - We present a family of kernels for analysis of data generated by dynamical systems these so-called cone kernels feature a dependence on the dynamical vector field operating in the phase space manifold, estimated empirically through finite differences of time-ordered data samples. In particular, cone kernels assign strong affinity to pairs of samples whose relative displacement vector lies within a narrow cone aligned with the dynamical vector field the outcome of this explicit dependence on the dynamics is that, in a suitable asymptotic limit, Laplace-Beltrami operators for data analysis constructed from cone kernels generate diffusions along the integral curves of the dynamical vector field. This property is independent of the observation modality, and endows these operators with invariance under a weakly restrictive class of transformations of the data (including conformal transformations), while it also enhances their capability to extract intrinsic dynamical timescales via eigenfunctions. Here, we study these features by establishing the Riemannian metric tensor induced by cone kernels in the limit of large data. We find that the corresponding Dirichlet energy is governed by the directional derivative of functions along the dynamical vector field, giving rise to a measure of roughness of functions that favors slowly varying observables. We demonstrate the utility of cone kernels in nonlinear flows on the 2-torus and North Pacific sea surface temperature data generated by a comprehensive climate model.

KW - Delay Coordinates

KW - Diffusion Operators

KW - Eigenfunctions

KW - Kernel Methods

KW - Manifold Embedding

KW - Vector Field

UR - http://www.scopus.com/inward/record.url?scp=84937849321&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937849321&partnerID=8YFLogxK

U2 - 10.1137/140954544

DO - 10.1137/140954544

M3 - Article

AN - SCOPUS:84937849321

VL - 14

SP - 556

EP - 608

JO - SIAM Journal on Applied Dynamical Systems

JF - SIAM Journal on Applied Dynamical Systems

SN - 1536-0040

IS - 2

ER -