Neural network research to better enable ethical AI decision makingNews
August 10, 2021
ABERDEEN PROVING GROUND, Md. Researchers from the U.S. Army Combat Capabilities Development Command, known as DEVCOM, Army Research Laboratory and university partners from the Internet of Battlefield Things Collaborative Research Alliance (IoBT CRA) announced the development of a method for neural networks to be more confident in their understanding of battlefield environments.
According to officials, researchers reviewed frameworks to represent uncertainty, categorized sources of uncertainty in military information-networks’ common operating environment, and then created solutions to manage uncertainty within systems.
The researchers developed insights from the uncertainty management approaches into a workflow that is intended to maximize effectiveness in accomplishing mission goals despite the presence of uncertainty in data inputs. Through this process, they are aiming to teach neural networks when to accurately be sure about a decision.
Researchers claim that this improved confidence in neural networks could have significant implications for the battlefield, as certainty in artificial intelligence (AI) decisions and behaviors are essential to ensure ethical and effective decision-making autonomy in combat.