We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Alignment and stability of embeddings: measurement and inference improvement

Abstract: Representation learning (RL) methods learn objects' latent embeddings where information is preserved by distances. Since distances are invariant to certain linear transformations, one may obtain different embeddings while preserving the same information. In dynamic systems, a temporal difference in embeddings may be explained by the stability of the system or by the misalignment of embeddings due to arbitrary transformations. In the literature, embedding alignment has not been defined formally, explored theoretically, or analyzed empirically. Here, we explore the embedding alignment and its parts, provide the first formal definitions, propose novel metrics to measure alignment and stability, and show their suitability through synthetic experiments. Real-world experiments show that both static and dynamic RL methods are prone to produce misaligned embeddings and such misalignment worsens the performance of dynamic network inference tasks. By ensuring alignment, the prediction accuracy raises by up to 90% in static and by 40% in dynamic RL methods.
Subjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Cite as: arXiv:2101.07251 [cs.LG]
  (or arXiv:2101.07251v1 [cs.LG] for this version)

Submission history

From: Furkan Gursoy [view email]
[v1] Mon, 18 Jan 2021 18:58:59 GMT (1208kb,D)

Link back to: arXiv, form interface, contact.