Floodgates up to contain the DeePC and limit extrapolation
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929690462126080 |
|---|---|
| author | Ramadan, Mohammad Toler, Evan Anitescu, Mihai |
| author_facet | Ramadan, Mohammad Toler, Evan Anitescu, Mihai |
| contents | Behavioral data-enabled control approaches typically assume data-generating systems of linear dynamics. This may result in false generalization if the newly designed closed-loop system results in input-output distributional shifts beyond learning data. These shifts may compromise safety by activating harmful nonlinearities in the data-generating system not experienced previously in the data and/or not captured by the linearity assumption inherent in these approaches. This paper proposes an approach to slow down the distributional shifts and therefore enhance the safety of the data-enabled methods. This is achieved by introducing quadratic regularization terms to the data-enabled predictive control formulations. Slowing down the distributional shifts comes at the expense of slowing down the exploration, in a trade-off resembling the exploration vs exploitation balance in machine learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_17318 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Floodgates up to contain the DeePC and limit extrapolation Ramadan, Mohammad Toler, Evan Anitescu, Mihai Systems and Control Behavioral data-enabled control approaches typically assume data-generating systems of linear dynamics. This may result in false generalization if the newly designed closed-loop system results in input-output distributional shifts beyond learning data. These shifts may compromise safety by activating harmful nonlinearities in the data-generating system not experienced previously in the data and/or not captured by the linearity assumption inherent in these approaches. This paper proposes an approach to slow down the distributional shifts and therefore enhance the safety of the data-enabled methods. This is achieved by introducing quadratic regularization terms to the data-enabled predictive control formulations. Slowing down the distributional shifts comes at the expense of slowing down the exploration, in a trade-off resembling the exploration vs exploitation balance in machine learning. |
| title | Floodgates up to contain the DeePC and limit extrapolation |
| topic | Systems and Control |
| url | https://arxiv.org/abs/2501.17318 |