On discovery and exploitation of temporal structure in data sets. / Scarfe, Tim.

2015. 175 p.

Research output: ThesisDoctoral Thesis

Unpublished

Standard

On discovery and exploitation of temporal structure in data sets. / Scarfe, Tim.

2015. 175 p.

Research output: ThesisDoctoral Thesis

Harvard

Scarfe, T 2015, 'On discovery and exploitation of temporal structure in data sets', Ph.D., Royal Holloway, University of London.

APA

Vancouver

Author

BibTeX

@phdthesis{e5e1b910f4f148548924ba10cf5f5a9a,
title = "On discovery and exploitation of temporal structure in data sets",
abstract = "This thesis explores temporal structure based on self-similarity in different contexts.An efficient dynamic programming algorithm is presented which discovers temporal structures in music shows, obtains high quality results, and compares them to similar algorithms used in the literature. The program segments a self-similarity matrix given a cost function and a fixed number of homogeneoustemporal structures to find. This is the initial approach we use to discover temporal structures in music data.The use of a self-similarity matrix to visualize temporal structures is discussed in detail. Then the following question is explored; if similar temporal structures in other corpora existed; could forecasting algorithms be adapted to take advantage of them even if they were not known a priori?Prediction with expert advice techniques are then introduced to exploit a priori unknown temporal structures of a similar configuration in an on-line configuration. Uni-variate Russian Stock Exchange options futures volatility corpora are used, which are highly interesting for on-line forecasting.We experiment with merging together expert models which have been trained in some way to recognise temporal structures in corpora. The first types are kernel ridge regression models trained to be experts on particular regions in time, oruntrained and given random sets of parameters which may work well on certain time regions. The other types of model used are parsimonious predictors which transform uni-variate financial data into elementary time series based on homogeneous vicinities of information in the side domain. Expert merging techniques are then used across these time series which produce a validation-free forecaster comparable to sliding kernel ridge regression.",
keywords = "Machine Learning, On-line regression, time series, music, dj, finance, forecasting, temporal structures, regions, regimes, prediction with expert advice, segmentation, merging, experts",
author = "Tim Scarfe",
year = "2015",
language = "English",
school = "Royal Holloway, University of London",

}

RIS

TY - THES

T1 - On discovery and exploitation of temporal structure in data sets

AU - Scarfe, Tim

PY - 2015

Y1 - 2015

N2 - This thesis explores temporal structure based on self-similarity in different contexts.An efficient dynamic programming algorithm is presented which discovers temporal structures in music shows, obtains high quality results, and compares them to similar algorithms used in the literature. The program segments a self-similarity matrix given a cost function and a fixed number of homogeneoustemporal structures to find. This is the initial approach we use to discover temporal structures in music data.The use of a self-similarity matrix to visualize temporal structures is discussed in detail. Then the following question is explored; if similar temporal structures in other corpora existed; could forecasting algorithms be adapted to take advantage of them even if they were not known a priori?Prediction with expert advice techniques are then introduced to exploit a priori unknown temporal structures of a similar configuration in an on-line configuration. Uni-variate Russian Stock Exchange options futures volatility corpora are used, which are highly interesting for on-line forecasting.We experiment with merging together expert models which have been trained in some way to recognise temporal structures in corpora. The first types are kernel ridge regression models trained to be experts on particular regions in time, oruntrained and given random sets of parameters which may work well on certain time regions. The other types of model used are parsimonious predictors which transform uni-variate financial data into elementary time series based on homogeneous vicinities of information in the side domain. Expert merging techniques are then used across these time series which produce a validation-free forecaster comparable to sliding kernel ridge regression.

AB - This thesis explores temporal structure based on self-similarity in different contexts.An efficient dynamic programming algorithm is presented which discovers temporal structures in music shows, obtains high quality results, and compares them to similar algorithms used in the literature. The program segments a self-similarity matrix given a cost function and a fixed number of homogeneoustemporal structures to find. This is the initial approach we use to discover temporal structures in music data.The use of a self-similarity matrix to visualize temporal structures is discussed in detail. Then the following question is explored; if similar temporal structures in other corpora existed; could forecasting algorithms be adapted to take advantage of them even if they were not known a priori?Prediction with expert advice techniques are then introduced to exploit a priori unknown temporal structures of a similar configuration in an on-line configuration. Uni-variate Russian Stock Exchange options futures volatility corpora are used, which are highly interesting for on-line forecasting.We experiment with merging together expert models which have been trained in some way to recognise temporal structures in corpora. The first types are kernel ridge regression models trained to be experts on particular regions in time, oruntrained and given random sets of parameters which may work well on certain time regions. The other types of model used are parsimonious predictors which transform uni-variate financial data into elementary time series based on homogeneous vicinities of information in the side domain. Expert merging techniques are then used across these time series which produce a validation-free forecaster comparable to sliding kernel ridge regression.

KW - Machine Learning

KW - On-line regression

KW - time series

KW - music

KW - dj

KW - finance

KW - forecasting

KW - temporal structures

KW - regions

KW - regimes

KW - prediction with expert advice

KW - segmentation

KW - merging

KW - experts

M3 - Doctoral Thesis

ER -