<ul data-eligibleForWebStory="false"><li>Researchers propose a method for verifying neural network policies in stochastic systems for reach-avoid specifications.</li><li>They introduce logarithmic Reach-Avoid Supermartingales (logRASMs) to achieve smaller Lipschitz constants than existing approaches.</li><li>A faster method to compute tighter upper bounds on Lipschitz constants based on weighted norms is presented in the study.</li><li>Empirical evaluation demonstrates successful verification of reach-avoid specifications with probabilities as high as 99.9999%.</li></ul>

Policy Verification in Stochastic Dynamical Systems Using Logarithmic Neural Certificates

Discover more