# Enhancing AI System Resiliency: Formulation and Guarantee for LSTM Resilience Based on Control Theory

arXiv: https://arxiv.org/abs/2505.17696 · HTML: https://mlsanity.com/paper/2505.17696/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 35 references that are in the map.
- First posted on arXiv 23 May 2025.

## Builds on (most cited first, 5 of 5)

- [Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions](https://mlsanity.com/paper/2109.06697/) (2021)
- [LSTM Neural Networks: Input to State Stability and Probabilistic Safety Verification](https://mlsanity.com/paper/1912.04377/) (2019)
- [How Robust are Deep Neural Networks?](https://mlsanity.com/paper/1804.11313/) (2018)
- [Learning of Multi-Context Models for Autonomous Underwater Vehicles](https://mlsanity.com/paper/1809.06179/) (2018)
- [Deep Long-Short Term Memory networks: Stability properties and Experimental validation](https://mlsanity.com/paper/2304.02975/) (2023)

## Abstract (arXiv)

This paper proposes a novel theoretical framework for guaranteeing and evaluating the resilience of long short-term memory (LSTM) networks in control systems. We introduce "recovery time" as a new metric of resilience in order to quantify the time required for an LSTM to return to its normal state after anomalous inputs. By mathematically refining incremental input-to-state stability (δISS) theory for LSTM, we derive a practical data-independent upper bound on recovery time. This upper bound gives us resilience-aware training. Experimental validation on simple models demonstrates the effectiveness of our resilience estimation and control methods, enhancing a foundation for rigorous quality assurance in safety-critical AI applications.