Oriel — AI & technology worth noticing, every day.

WETW: Learning Pair-Specific Weighted Euclidean Distances for Efficient DTW Search

Paper · https://arxiv.org/abs/2611.17412

arXiv:2611.17412v1 Announce Type: new Abstract: Dynamic Time Warping (DTW) is the standard distance for similarity search over time series that are misaligned in time, but its cost is quadratic in the series length and lower bounds often prune few candidates, so search over large collections of long series is slow. Learned approximations embed the series to make comparisons cheap, yet compare every pair with the same fixed rule, usually the Euclidean distance, although DTW aligns each pair differently. We observe that, once a pair is aligned, DTW is a Euclidean distance over coordinates selected for that pair. Building on this view, we propose Weighted Euclidean Time Warping (WETW), a learned distance that adapts the comparison of fixed-length embeddings to each pair. A small symmetric network predicts one weight per embedding di- mension and pair without computing an alignment, which halves the error of the best static distance on the same embeddings. The embeddings come from the Weighted Euclidean SEries Encoder (WESEE), whose multi-scale Temporal Feature Extractor keeps the timing information that global pooling removes and raises recall@1 from 0.90 to 0.95, although WETW works with any encoder. We also show that fitting DTW distances is not enough for search, as regression alone misorders the closest neighbours, and that ranking regularization raises recall@1 from 0.79 to 0.95. Our filter-and-verify pipeline, WETW-FV, uses WETW to build a shortlist and exact DTW to check it, so the search is approximate but every returned distance is exact. Over eight datasets, WETW-FV is 11.4x faster than the fastest exact method among the UCR Suite, lower bounds and MESSI-DTW, at a mean recall@10 of 0.93, while the best learned baseline reaches 0.59 at the same speed. On 48K-point series, WETW- FV remains 40x faster than the UCR Suite at a recall@10 of 0.91.

First seen by Oriel on .

AI summary

This paper introduces WETW, a pair-adaptive weighted Euclidean distance for learned time-series embeddings, and reports faster approximate DTW search with exact verification and substantially higher recall than learned baselines.

  • WETW uses a small symmetric network to predict one weight per embedding dimension for each pair without computing an alignment.
  • WESEE’s multi-scale Temporal Feature Extractor preserves timing information, increasing recall@1 from 0.90 to 0.95; WETW also supports other encoders.
  • Regression alone misorders nearest neighbours, while ranking regularization improves recall@1 from 0.79 to 0.95.
  • Across eight datasets, WETW-FV is 11.4x faster than the fastest exact method, with mean recall@10 of 0.93 versus 0.59 for the best learned baseline.
  • On 48K-point series, WETW-FV remains 40x faster than the UCR Suite at recall@10 of 0.91, while exact DTW verifies returned distances.

#dynamic time warping #time series #similarity search #metric learning

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
arXiv AI Papersoai:arXiv.org:2611.17412v1https://arxiv.org/abs/2611.17412 Read discussion