Data-driven AI/ML techniques have advanced significantly to automate skills such as detection, target recognition, and mobility. Yet, there are many applications, such as military operation or humanitarian assistance and disaster relief, where it is highly likely that the operating domain will depict some distributional shift from that in which a system was trained. Under these scenarios, the design of AI systems that can be trained or refined quickly, potentially in real-time, becomes critically important to ensure safety and success. I will discuss how we are incorporating learning from human demonstration to address the need for efficient, on-line learning in the field. I will specifically focus on several approaches to learn navigation behaviors for unmanned ground robots using teleoperation demonstrations, allowing for non-expert users to refine ML reward functions with relatively little effort. Maggie Wigness Computer Scientist Combat Capabilities Development Command Army Research Laboratory To ask the speaker a question, click on the speech bubble icon in the lower right hand corner and type in the question in the window that pops up. The question will be sent directly to us. Please note that there is a little bit of a delay when streaming. What participants see is a few minutes behind what is happening at our end. The longer we stream, the greater the delay may become so the questions submitted at the very end may not reach us in time. The best way to get the questions answered is to send them as they come up.