Abstract:The proliferation and ubiquity of temporal data across many disciplines has sparked interest for similarity, classification and clustering methods specifically designed to handle time series data. A core issue when dealing with time series is determining their pairwise similarity, i.e., the degree to which a given time series resembles another. Traditional distance measures such as the Euclidean are not well-suited due to the time-dependent nature of the data. Elastic metrics such as dynamic time warping (DTW) offer a promising approach, but are limited by their computational complexity, non-differentiability and sensitivity to noise and outliers. This thesis proposes novel elastic alignment methods that use parametric \& diffeomorphic warping transformations as a means of overcoming the shortcomings of DTW-based metrics. The proposed method is differentiable \& invertible, well-suited for deep learning architectures, robust to noise and outliers, computationally efficient, and is expressive and flexible enough to capture complex patterns. Furthermore, a closed-form solution was developed for the gradient of these diffeomorphic transformations, which allows an efficient search in the parameter space, leading to better solutions at convergence. Leveraging the benefits of these closed-form diffeomorphic transformations, this thesis proposes a suite of advancements that include: (a) an enhanced temporal transformer network for time series alignment and averaging, (b) a deep-learning based time series classification model to simultaneously align and classify signals with high accuracy, (c) an incremental time series clustering algorithm that is warping-invariant, scalable and can operate under limited computational and time resources, and finally, (d) a normalizing flow model that enhances the flexibility of affine transformations in coupling and autoregressive layers.

Similarity-Based Queries for Time Series Data

Dynamic Time Warping under Product Quantization, with Applications to Time-Series Data Similarity Search

An efficient method for time series similarity search using binary code representation and hamming distance

A New Representation and Distance Measure for Financial Time Series

A Novel Similarity Measure Approach for Time Series Based on PLA and DTW

A Novel Similarity Measure Model for Multivariate Time Series Based on LMNN and DTW

Micro Similarity Queries in Time Series Database

Interest-Based Queries For Time Series Data

A Framework for Similarity Search of Time Series Cliques with Natural Relations

Towards a faster symbolic aggregate approximation method

A New Method for Similarity Matching of Non-Stationary Time Series Based on Fractal Time-Varying Dimension

Return of the Lernaean Hydra: Experimental Evaluation of Data Series Approximate Similarity Search

Efficient Algorithm for a Novel Pattern of Time Series

Piecewise Statistic Approximation Based Similarity Measure for Time Series

Subsequence Similarity Search under Time Shifting

Diffeomorphic Transformations for Time Series Analysis: An Efficient Approach to Nonlinear Warping

Fast Online Similarity Search for Uncertain Time Series

Efficient Similarity Matching of Time Series Cliques with Natural Relations

Continually Evaluating Similarity-Based Pattern Queries on a Streaming Time Series

Similarity Join on Time Series by Utilizing a Dynamic Segmentation Index

Similarity Search in Trajectory Databases