NExG: Provable and Guided State Space Exploration of Neural Network Control Systems using Sensitivity Approximation

Autor:	Goyal, Manish, Dewaskar, Miheer, Duggirala, Parasara Sridhar
Rok vydání:	2022
Předmět:	Electrical Engineering and Systems Science - Systems and Control Computer Science - Machine Learning Computer Science - Robotics 93-04 93-08 93-11 68T07 68U07 F.4 I.6 J.6 G.4
Druh dokumentu:	Working Paper
Popis:	We propose a new technique for performing state space exploration of closed loop control systems with neural network feedback controllers. Our approach involves approximating the sensitivity of the trajectories of the closed loop dynamics. Using such an approximator and the system simulator, we present a guided state space exploration method that can generate trajectories visiting the neighborhood of a target state at a specified time. We present a theoretical framework which establishes that our method will produce a sequence of trajectories that will reach a suitable neighborhood of the target state. We provide thorough evaluation of our approach on various systems with neural network feedback controllers of different configurations. We outperform earlier state space exploration techniques and achieve significant improvement in both the quality (explainability) and performance (convergence rate). Finally, we adopt our algorithm for the falsification of a class of temporal logic specification, assess its performance against a state-of-the-art falsification tool, and show its potential in supplementing existing falsification algorithms. Comment: 13 pages, 9 figures, 4 tables. Accepted at EMSOFT 2022 for publication in IEEE TCAD
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2207.03884 Zobrazit plný text záznamu View this record from Arxiv