Uniform Error Bounds for Gaussian Process Regression with Application to Safe Control
Armin Lederer
Abstract
Motivating Example Nonparameteric regresssion offers great promises in robotic applicationsLearned policies are unsafe in real world applications [1] Constrained environments to avoid damages of hardware No human-robot interaction due to risk of injuries START GOAL Quantification of uncertainty in data-driven models essential for safety-critical applications Robust control for rigorous safety certificates How can the learning error be bounded based on the model uncertainty?How are formal safety guarantees provided for policies based on uncertain models?
Topics & Concepts
Lipschitz continuityGaussian processGaussianComputer scienceProbabilistic logicMeasure (data warehouse)KrigingApplied mathematicsProcess (computing)Key (lock)Upper and lower boundsDynamical systems theoryMathematicsMathematical optimizationAlgorithmArtificial intelligenceData miningMachine learningMathematical analysisOperating systemQuantum mechanicsComputer securityPhysicsGaussian Processes and Bayesian InferenceAdvanced Control Systems OptimizationFault Detection and Control Systems