Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2401.16808 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913463739088896 |
|---|---|
| author | Choi, Insu Koh, Woosung Kang, Gimin Jang, Yuntae Kim, Woo Chang |
| author_facet | Choi, Insu Koh, Woosung Kang, Gimin Jang, Yuntae Kim, Woo Chang |
| contents | Modeling time series data remains a pervasive issue as the temporal dimension is inherent to numerous domains. Despite significant strides in time series forecasting, high noise-to-signal ratio, non-normality, non-stationarity, and lack of data continue challenging practitioners. In response, we leverage a simple representation augmentation technique to overcome these challenges. Our augmented representation acts as a statistical-space prior encoded at each time step. In response, we name our method Statistical-space Augmented Representation (SSAR). The underlying high-dimensional data-generating process inspires our representation augmentation. We rigorously examine the empirical generalization performance on two data sets with two downstream temporal learning algorithms. Our approach significantly beats all five up-to-date baselines. Moreover, the highly modular nature of our approach can easily be applied to various settings. Lastly, fully-fledged theoretical perspectives are available throughout the writing for a clear and rigorous understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16808 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Encoding Temporal Statistical-space Priors via Augmented Representation Choi, Insu Koh, Woosung Kang, Gimin Jang, Yuntae Kim, Woo Chang Machine Learning Artificial Intelligence Modeling time series data remains a pervasive issue as the temporal dimension is inherent to numerous domains. Despite significant strides in time series forecasting, high noise-to-signal ratio, non-normality, non-stationarity, and lack of data continue challenging practitioners. In response, we leverage a simple representation augmentation technique to overcome these challenges. Our augmented representation acts as a statistical-space prior encoded at each time step. In response, we name our method Statistical-space Augmented Representation (SSAR). The underlying high-dimensional data-generating process inspires our representation augmentation. We rigorously examine the empirical generalization performance on two data sets with two downstream temporal learning algorithms. Our approach significantly beats all five up-to-date baselines. Moreover, the highly modular nature of our approach can easily be applied to various settings. Lastly, fully-fledged theoretical perspectives are available throughout the writing for a clear and rigorous understanding. |
| title | Encoding Temporal Statistical-space Priors via Augmented Representation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2401.16808 |