Asynchronous Methods for Deep Reinforcement Learning: MuJoCo
The video shows agents trained using the Asynchronous Advantage Actor-Critic (A3C) algorithm performing a variety of motor control tasks. The tasks successfully learned by the agents include pole swing-up, quadruped locomotion, planar biped walking, balancing, 2D target reaching, and 3D manipulation. Paper link - http://arxiv.org/pdf/1602.01783.pdf
Видео Asynchronous Methods for Deep Reinforcement Learning: MuJoCo канала DeepMind
Видео Asynchronous Methods for Deep Reinforcement Learning: MuJoCo канала DeepMind
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Google DeepMind AI Does Parkour - Producing flexible behaviours in simulated environments
Robot Control with Distributed Deep Reinforcement Learning
MarI/O - Machine Learning for Video Games
TRPO MuJoCo
Deep Learning Cars
Reinforcement learning for a robitic soccer goalkeeper
Autonomous robot car control demonstration in CES2016
Robot Control with Reinforcement Learning and Neural Network
AI learns to play snake using Genetic Algorithm and Deep learning
Deep reinforcement learning for Flappy Bird
Flappy Bird AI with A3C
Efficient Reinforcement Learning for Robots using Informative Simulated Priors
Google DeepMind's Deep Q-learning playing Atari Breakout
Why Concrete Needs Reinforcement
Google's DeepMind AI Just Taught Itself To Walk
Reinforcement learning policy for 30 legged centipede in Mujoco
DeepMind Lab - Stairway to Melon Level
Learning to move: DDPG Algorithm on Gym MuJoCo
Deep reinforcement learning with intrinsic motivation and temporal abstractions
DDPG mujoco sim reacher