acm-header
Sign In

Communications of the ACM

ACM Careers

Computer Scientists Develop System That Helps AI Understand Human Goals


human hand and robotic figures in human-robot interaction illustration

The framework enables in situ bidirectional human-robot value alignment.

Credit: Zhen Chen / BIGAI

Researchers worldwide are racing to develop AI systems that can communicate with their human operators in a language they can understand. One team has published a paper in the journal Science Robotics that presents a framework for what they call "explainable artificial intelligence," or XAI. 

The researchers, from the U.S. and China, show that real-time human-robot mutual understanding in complex cooperative tasks is achievable with a learning model based on bidirectional communication.

From Interesting Engineering
View Full Article – May Require Paid Registration


 

No entries found