Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2404.12030 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929318750322688 |
|---|---|
| author | Drummond, Ross Baldivieso-Monasterios, Pablo R Valmorbida, Giorgio |
| author_facet | Drummond, Ross Baldivieso-Monasterios, Pablo R Valmorbida, Giorgio |
| contents | Model predictive control (MPC) for linear systems with quadratic costs and linear constraints is shown to admit an exact representation as an implicit neural network. A method to "unravel" the implicit neural network of MPC into an explicit one is also introduced. As well as building links between model-based and data-driven control, these results emphasize the capability of implicit neural networks for representing solutions of optimisation problems, as such problems are themselves implicitly defined functions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_12030 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Mapping back and forth between model predictive control and neural networks Drummond, Ross Baldivieso-Monasterios, Pablo R Valmorbida, Giorgio Systems and Control Artificial Intelligence Model predictive control (MPC) for linear systems with quadratic costs and linear constraints is shown to admit an exact representation as an implicit neural network. A method to "unravel" the implicit neural network of MPC into an explicit one is also introduced. As well as building links between model-based and data-driven control, these results emphasize the capability of implicit neural networks for representing solutions of optimisation problems, as such problems are themselves implicitly defined functions. |
| title | Mapping back and forth between model predictive control and neural networks |
| topic | Systems and Control Artificial Intelligence |
| url | https://arxiv.org/abs/2404.12030 |