Multiple-Resolution Tokenization for Time Series Forecasting with an Application to Pricing
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917991254327296 |
|---|---|
| author | Peršak, Egon Anjos, Miguel F. Lautz, Sebastian Kolev, Aleksandar |
| author_facet | Peršak, Egon Anjos, Miguel F. Lautz, Sebastian Kolev, Aleksandar |
| contents | We propose a transformer architecture for time series forecasting with a focus on time series tokenisation and apply it to a real-world prediction problem from the pricing domain. Our architecture aims to learn effective representations at many scales across all available data simultaneously. The model contains a number of novel modules: a differentiated form of time series patching which employs multiple resolutions, a multiple-resolution module for time-varying known variables, a mixer-based module for capturing cross-series information, and a novel output head with favourable scaling to account for the increased number of tokens. We present an application of this model to a real world prediction problem faced by the markdown team at a very large retailer. On the experiments conducted our model outperforms in-house models and the selected existing deep learning architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_03185 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Multiple-Resolution Tokenization for Time Series Forecasting with an Application to Pricing Peršak, Egon Anjos, Miguel F. Lautz, Sebastian Kolev, Aleksandar Machine Learning Artificial Intelligence We propose a transformer architecture for time series forecasting with a focus on time series tokenisation and apply it to a real-world prediction problem from the pricing domain. Our architecture aims to learn effective representations at many scales across all available data simultaneously. The model contains a number of novel modules: a differentiated form of time series patching which employs multiple resolutions, a multiple-resolution module for time-varying known variables, a mixer-based module for capturing cross-series information, and a novel output head with favourable scaling to account for the increased number of tokens. We present an application of this model to a real world prediction problem faced by the markdown team at a very large retailer. On the experiments conducted our model outperforms in-house models and the selected existing deep learning architectures. |
| title | Multiple-Resolution Tokenization for Time Series Forecasting with an Application to Pricing |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2407.03185 |