Recovering Formal Guarantees for Neural Network-Controlled Systems Under Changing Dynamics
Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Doctoral Consortium. Pages 8335-8336.
https://doi.org/10.24963/ijcai.2026/946
Keywords
- Uncertainty in AI: Sequential decision making
- Agent-based and Multi-agent Systems: Formal verification, validation and synthesis
- AI Ethics, Trust, Fairnes: Safety and robustness
Abstract
Neural networks enable high-performance decision-making, but their deployment in safety-critical settings requires formal guarantees that may become invalid after changes in the system dynamics. This research studies how such guarantees can be recovered using information from the initial verification, enabling post-deployment analysis to assess preservation, quantify degradation, and restore compliant behavior without recomputing verification from scratch.
@inproceedings{ijcai2026p946,
title = {Recovering Formal Guarantees for Neural Network-Controlled Systems Under Changing Dynamics},
author = {Lutz, Sterre},
booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI-26}},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
editor = {Diego Calvanese},
pages = {8335--8336},
year = {2026},
month = {8},
note = {Doctoral Consortium},
doi = {10.24963/ijcai.2026/946},
url = {https://doi.org/10.24963/ijcai.2026/946},
}
TY - CPAPER TI - Recovering Formal Guarantees for Neural Network-Controlled Systems Under Changing Dynamics AU - Lutz, Sterre T2 - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26 PB - International Joint Conferences on Artificial Intelligence Organization PY - 2026 SP - 8335 EP - 8336 DO - 10.24963/ijcai.2026/946 UR - https://doi.org/10.24963/ijcai.2026/946 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
