DualCast: A Model to Disentangle Aperiodic Events from Traffic Series
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910993700880384 |
|---|---|
| author | Su, Xinyu Liu, Feng Chang, Yanchuan Tanin, Egemen Sarvi, Majid Qi, Jianzhong |
| author_facet | Su, Xinyu Liu, Feng Chang, Yanchuan Tanin, Egemen Sarvi, Majid Qi, Jianzhong |
| contents | Traffic forecasting is crucial for transportation systems optimisation. Current models minimise the mean forecasting errors, often favouring periodic events prevalent in the training data, while overlooking critical aperiodic ones like traffic incidents. To address this, we propose DualCast, a dual-branch framework that disentangles traffic signals into intrinsic spatial-temporal patterns and external environmental contexts, including aperiodic events. DualCast also employs a cross-time attention mechanism to capture high-order spatial-temporal relationships from both periodic and aperiodic patterns. DualCast is versatile. We integrate it with recent traffic forecasting models, consistently reducing their forecasting errors by up to 9.6% on multiple real datasets. Our source code is available at https://github.com/suzy0223/DualCast. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_18286 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DualCast: A Model to Disentangle Aperiodic Events from Traffic Series Su, Xinyu Liu, Feng Chang, Yanchuan Tanin, Egemen Sarvi, Majid Qi, Jianzhong Machine Learning Artificial Intelligence Traffic forecasting is crucial for transportation systems optimisation. Current models minimise the mean forecasting errors, often favouring periodic events prevalent in the training data, while overlooking critical aperiodic ones like traffic incidents. To address this, we propose DualCast, a dual-branch framework that disentangles traffic signals into intrinsic spatial-temporal patterns and external environmental contexts, including aperiodic events. DualCast also employs a cross-time attention mechanism to capture high-order spatial-temporal relationships from both periodic and aperiodic patterns. DualCast is versatile. We integrate it with recent traffic forecasting models, consistently reducing their forecasting errors by up to 9.6% on multiple real datasets. Our source code is available at https://github.com/suzy0223/DualCast. |
| title | DualCast: A Model to Disentangle Aperiodic Events from Traffic Series |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2411.18286 |