Reinforcement Learning with Prediction-Based Rewards
We’ve developed Random Network Distillation (RND), a prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity, which for the first time exceeds average human performance on Montezuma’s Revenge. Learn more: https://blog.openai.com/reinforcement-learning-with-prediction-based-rewards/
Видео Reinforcement Learning with Prediction-Based Rewards канала OpenAI
Видео Reinforcement Learning with Prediction-Based Rewards канала OpenAI
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Neural Network Learns to Play SnakeDeep Learning CarsMontezuma's Revenge 11300 Scores with OpenAI Random Network Distillation By 4 Titan XpMulti-Agent Hide and SeekMeta-Reinforcement Learning for Robotic Industrial Insertion TasksThis AI Senses Humans Through Walls 👀MarI/O - Machine Learning for Video GamesGoogle DeepMind's Deep Q-learning playing Atari BreakoutGoogle's Deep Mind Explained! - Self Learning A.I.AI learns to play snake using Genetic Algorithm and Deep learningMachine Learning TetrisDeep Q Network learning to play Galaxian (cleaner encode)Deep Reinforcement Learning in Pac-manIntrinsically Motivated Goal Exploration, Automated Curriculum Learning and Emergence of Tool UseReinforcement Learning Tutorial | Reinforcement Learning Example Using Python | EdurekaOperant conditioning: Positive-and-negative reinforcement and punishment | MCAT | Khan AcademyAI Learns to Park - Deep Reinforcement LearningCuriosity-Driven Learning: AI agents exploring without looking at any scoresReinforcement Learning in practice. AI playing games. Breakout