My research focuses on improving the ability of artificial agents to cope with circumstances outside of what was seen during an initial training phase.