Dynamic Boundary Time Warping for Sub-sequence Matching with Few Examples

TL;DR

This paper introduces a new dynamic boundary time warping method for sub-sequence matching that effectively handles few-shot query examples without averaging, outperforming or matching existing approaches in NLP tasks.

Contribution

It proposes a novel DTW-based algorithm for few-shot sub-sequence retrieval that does not require averaging query examples, enhancing accuracy in low-data scenarios.

Findings

Outperforms baseline methods in NLP tasks with few examples

Achieves comparable results to existing methods with limited data

Demonstrates effectiveness of direct query usage without averaging

Abstract

The paper presents a novel method of finding a fragment in a long temporal sequence similar to the set of shorter sequences. We are the first to propose an algorithm for such a search that does not rely on computing the average sequence from query examples. Instead, we use query examples as is, utilizing all of them simultaneously. The introduced method based on the Dynamic Time Warping (DTW) technique is suited explicitly for few-shot query-by-example retrieval tasks. We evaluate it on two different few-shot problems from the field of Natural Language Processing. The results show it either outperforms baselines and previous approaches or achieves comparable results when a low number of examples is available.