Researchers from the University of Illinois at Urbana-Champaign (USA) have developed an artificial intelligence (AI)-based method for teaching robots to work together.
This was reported by ScienceDaily, according to Ukrinform reports.
“We have developed a machine learning-based method that allows us to identify when a single machine contributes to a team goal. If you look at the sports field, a football player can score a goal, but we want to know the actions of other teammates that made the goal, such as who provided the assist,” said researcher Hue. tran. .
A reinforcement learning algorithm may also detect that the robot is doing something that does not contribute to the goal.
According to Tan, such an algorithm can be applied to many real-world situations, such as military surveillance, warehouse operations, traffic light management, autonomous vehicles coordinating deliveries, or power grid monitoring.
Reinforcement learning is a field of machine learning inspired by behavioral psychology that studies what actions programs must perform in a given environment in order to maximize the concept of total reward.