We describe and evaluate LF-checker, a metaverifier tool based on machine learning. It extracts multiple features of the program under test and predicts the optimal configuration (flags) of a bounded model checker with a decision tree. Our current work is specialised in concurrency verification and employs ESBMC as a back-end verification engine. In the paper, we demonstrate that LF-checker achieves better results than the default configuration of the underlying verification engine.

In 2020, poor-quality software systems led to financial losses of approximately USD 2.08 trillion in the U.S. alone.19 Formal methods, such as bounded model checking (BMC), help to improve software quality, but they often fail to scale to the size and complexity of software.

Temporal logics can be used in reasoning about actions for specifying constraints on domain descriptions and temporal properties to be verified. In this paper, we exploit Bounded Model Checking (BMC) techniques in the verification of Dynamic Linear Time Temporal Logic (DLTL) properties of an action theory, which is formulated in a temporal extension of Answer Set Programming (ASP). To achieve completeness, we propose an approach to BMC which exploits the Buechi automaton construction while searching for a counterexample. We provide an encoding in ASP of the temporal action domain and of Bounded Model Checking of DLTL formulas.

At the University of Washington (UW) Medical Center, a radiotherapy system shoots high-powered radiation beams into the heads of patients, to treat cancers of the tongue and esophagus. Any software errors in the system could prove fatal, so engineers at the medical center have teamed with a group of computer scientists from the university to ensure the system will not fail, and that the beam will shut off if prescribed settings go out of tolerance. This is made possible by a process known as software verification, and verifying implementations of critical systems like that radiotherapy setup is one of the things about which Zachary Tatlock is passionate. Over three years ago, Tatlock was a Ph.D. candidate giving a talk at the university on his thesis research in program verification. The lead engineer for the medical center's radiotherapy team was in the audience, and asked Tatlock how they could apply verification to that system.