首页|Researchers at Suzhou University of Science and Technology Release New Data on R obotics (Reset-free Reinforcement Learning Via Multi-state Recovery and Failure Prevention for Autonomous Robots)
Researchers at Suzhou University of Science and Technology Release New Data on R obotics (Reset-free Reinforcement Learning Via Multi-state Recovery and Failure Prevention for Autonomous Robots)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Robotics have been pr esented. According to news reporting originating in Suzhou, People's Republic of China, by NewsRx journalists, research stated, "Reinforcement learning holds pr omise in enabling robotic tasks as it can learn optimal policies via trial and e rror. However, the practical deployment of reinforcement learning usually requir es human intervention to provide episodic resets when a failure occurs." Financial support for this research came from National Natural Science Foundatio n of China (NSFC). The news reporters obtained a quote from the research from the Suzhou University of Science and Technology, "Since manual resets are generally unavailable in au tonomous robots, we propose a reset-free reinforcement learning algorithm based on multi-state recovery and failure prevention to avoid failureinduced resets. The multi-state recovery provides robots with the capability of recovering from failures by self-correcting its behavior in the problematic state and, more impo rtantly, deciding which previous state is the best to return to for efficient re -learning. The failure prevention reduces potential failures by predicting and e xcluding possible unsafe actions in specific states."
SuzhouPeople's Republic of ChinaAsiaAutonomous RobotEmerging TechnologiesMachine LearningNano-robotReinfor cement LearningRoboticsSuzhou University of Science and Technology
NETL
NSTL
