Researchers at the US Army Research Laboratory (ARL) and the University of Texas (UT) in Austin have developed new robotic or computer programming techniques to learn how to perform tasks by interacting with human instructors. The results of this study were announced at the Artificial Intelligence Promotion Association Forum in New Orleans, Louisiana from February 2nd to 7th. ARL and UT researchers have considered a specific situation where people provide real-time feedback in the form of reviews. Dr. Peter Stone, a co-author of the University of Texas at Austin, and his predecessor, Dr. Brad Knox, developed TAMER first, or through evaluation, enhanced the manual training of robots. The ARL/UT team developed a new algorithm called Deep TAMER. This is an extension of TAMER, which uses deep learning. It is inspired by the human brain and allows robots to gain the ability to perform tasks by watching videos for a short time. According to Dr. Garrett Warnell, a military researcher, the research team believes that a person can teach robots how to do things by observing and providing comments (for example, "good job" or "bad job") just as humans train dogs to do tricks. Warnell said that researchers have extended the early work in this field and trained this type of robot or computer program to look at the world through images. This is an important first step in designing learning agents that can operate in the real world. External Cup Headsets,Threaded Bicycle Headsets,External Cup Bicycle Headsets,Easy Installation Cup Shenzhen Gineyea Technology Co., LTD. , https://www.gineyea.com