UrbanAI 2025 Challenge: Linear vs Transformer Models for Long-Horizon Exogenous Temperature Forecasting
Fuente:
arXiv
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915669757394944 |
|---|---|
| author | Gokhman, Ruslan |
| author_facet | Gokhman, Ruslan |
| contents | We study long-horizon exogenous-only temperature forecasting - a challenging univariate setting where only the past values of the indoor temperature are used for prediction - using linear and Transformer-family models. We evaluate Linear, NLinear, DLinear, Transformer, Informer, and Autoformer under standardized train, validation, and test splits. Results show that linear baselines (Linear, NLinear, DLinear) consistently outperform more complex Transformer-family architectures, with DLinear achieving the best overall accuracy across all splits. These findings highlight that carefully designed linear models remain strong baselines for time series forecasting in challenging exogenous-only settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_10866 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | UrbanAI 2025 Challenge: Linear vs Transformer Models for Long-Horizon Exogenous Temperature Forecasting Gokhman, Ruslan Machine Learning Artificial Intelligence We study long-horizon exogenous-only temperature forecasting - a challenging univariate setting where only the past values of the indoor temperature are used for prediction - using linear and Transformer-family models. We evaluate Linear, NLinear, DLinear, Transformer, Informer, and Autoformer under standardized train, validation, and test splits. Results show that linear baselines (Linear, NLinear, DLinear) consistently outperform more complex Transformer-family architectures, with DLinear achieving the best overall accuracy across all splits. These findings highlight that carefully designed linear models remain strong baselines for time series forecasting in challenging exogenous-only settings. |
| title | UrbanAI 2025 Challenge: Linear vs Transformer Models for Long-Horizon Exogenous Temperature Forecasting |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2512.10866 |