Minimizing Execution Duration in the Presence of Learning-Enabled Components
Kunal Agrawala, Sanjoy Baruahb, Alan Burnsc and Abhishek Singhd
akunal@wustl.edu
bbaruah@wustl.edu
calan.burns@york.ac.uk
dabhishek.s@wustl.edu
ABSTRACT
Autonomous systems are increasingly using components that incorporate machine learning and other AI-based techniques in order to achieve improved performance. We address the problem of assuring correctness in safety-critical systems that use such components. We investigate an approach which formulates the problem as one in which performance is an objective function to be optimized while safety is a hard constraint that must be satisfied. We then apply heuristics and algorithmic techniques from optimization theory in order to solve the resulting constrained optimization problem.
Keywords: Learning-enabled components (LECs), Safety critical systems, Performance optimization, Run-time monitoring, Typical analysis.
