TSRM: A Lightweight Temporal Feature Encoding Architecture for Time Series Forecasting and Imputation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913809990418432 |
|---|---|
| author | Leppich, Robert Stenger, Michael Grillmeyer, Daniel Borst, Vanessa Kounev, Samuel |
| author_facet | Leppich, Robert Stenger, Michael Grillmeyer, Daniel Borst, Vanessa Kounev, Samuel |
| contents | We introduce a temporal feature encoding architecture called Time Series Representation Model (TSRM) for multivariate time series forecasting and imputation. The architecture is structured around CNN-based representation layers, each dedicated to an independent representation learning task and designed to capture diverse temporal patterns, followed by an attention-based feature extraction layer and a merge layer, designed to aggregate extracted features. The architecture is fundamentally based on a configuration that is inspired by a Transformer encoder, with self-attention mechanisms at its core. The TSRM architecture outperforms state-of-the-art approaches on most of the seven established benchmark datasets considered in our empirical evaluation for both forecasting and imputation tasks. At the same time, it significantly reduces complexity in the form of learnable parameters. The source code is available at https://github.com/RobertLeppich/TSRM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_18878 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TSRM: A Lightweight Temporal Feature Encoding Architecture for Time Series Forecasting and Imputation Leppich, Robert Stenger, Michael Grillmeyer, Daniel Borst, Vanessa Kounev, Samuel Machine Learning Artificial Intelligence We introduce a temporal feature encoding architecture called Time Series Representation Model (TSRM) for multivariate time series forecasting and imputation. The architecture is structured around CNN-based representation layers, each dedicated to an independent representation learning task and designed to capture diverse temporal patterns, followed by an attention-based feature extraction layer and a merge layer, designed to aggregate extracted features. The architecture is fundamentally based on a configuration that is inspired by a Transformer encoder, with self-attention mechanisms at its core. The TSRM architecture outperforms state-of-the-art approaches on most of the seven established benchmark datasets considered in our empirical evaluation for both forecasting and imputation tasks. At the same time, it significantly reduces complexity in the form of learnable parameters. The source code is available at https://github.com/RobertLeppich/TSRM. |
| title | TSRM: A Lightweight Temporal Feature Encoding Architecture for Time Series Forecasting and Imputation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2504.18878 |