Approximately Equivariant Recurrent Generative Models for Quasi-Periodic Time Series with a Progressive Training Scheme

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fulek, Ruwen, Lange-Hegermann, Markus
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910114051522560
author Fulek, Ruwen
Lange-Hegermann, Markus
author_facet Fulek, Ruwen
Lange-Hegermann, Markus
contents We present a simple yet effective generative model for time series, based on a Recurrent Variational Autoencoder that we refer to as AEQ-RVAE-ST. Recurrent layers often struggle with unstable optimization and poor convergence when modeling long sequences. To address these limitations, we introduce a training scheme that subsequently increases the sequence length, stabilizing optimization and enabling consistent learning over extended horizons. By composing known components into a recurrent, approximately time-shift-equivariant topology, our model introduces an inductive bias that aligns with the structure of quasi-periodic and nearly stationary time series. Across several benchmark datasets, AEQ-RVAE-ST matches or surpasses state-of-the-art generative models, particularly on quasi-periodic data, while remaining competitive on more irregular signals. Performance is evaluated through ELBO, Fréchet Distance, discriminative metrics, and visualizations of the learned latent embeddings.
format Preprint
id arxiv_https___arxiv_org_abs_2505_05020
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Approximately Equivariant Recurrent Generative Models for Quasi-Periodic Time Series with a Progressive Training Scheme
Fulek, Ruwen
Lange-Hegermann, Markus
Machine Learning
We present a simple yet effective generative model for time series, based on a Recurrent Variational Autoencoder that we refer to as AEQ-RVAE-ST. Recurrent layers often struggle with unstable optimization and poor convergence when modeling long sequences. To address these limitations, we introduce a training scheme that subsequently increases the sequence length, stabilizing optimization and enabling consistent learning over extended horizons. By composing known components into a recurrent, approximately time-shift-equivariant topology, our model introduces an inductive bias that aligns with the structure of quasi-periodic and nearly stationary time series. Across several benchmark datasets, AEQ-RVAE-ST matches or surpasses state-of-the-art generative models, particularly on quasi-periodic data, while remaining competitive on more irregular signals. Performance is evaluated through ELBO, Fréchet Distance, discriminative metrics, and visualizations of the learned latent embeddings.
title Approximately Equivariant Recurrent Generative Models for Quasi-Periodic Time Series with a Progressive Training Scheme
topic Machine Learning
url https://arxiv.org/abs/2505.05020