There is a growing trend towards autonomy in present and forthcoming computing applications, including web-services and autonomous vehicles. Many of these applications are based on the concept of autonomous agent.
The group focuses on developing methods aimed at verifying that autonomous multi-agent systems meet their specifications. Specifically, the group is concerned with developing efficient model checking techniques and tools to verify multi-agent systems specified by agent-based logics. The research draws from areas such as modal logic, multi-agent systems, and model checking.
More recently, systems based on neural networks have become of increasing importance. The group is now also investigating techniques to allow verification of systems based on neural networks. This research is partly funded by DARPA’s Assured Autonomy program.
We are always looking for passionate new PhD students, Postdocs, and Master students to join the team (more info) !
Formal Verification of Neural Agents in Non-deterministic Environments
2nd March 2020, 12pm
Room 217, Huxley Building