Learning long-horizon behaviors, from latent imagination images, the Google Dreamer reinforcement learning agent takes fewer episodes, than strong model free agents, to solve visual control tasks.
Hunted by @nigelmarkdias
Learning long-horizon behaviors, from latent imagination images, the Google Dreamer reinforcement learning agent takes fewer episodes, than strong model free agents, to solve visual control tasks.
$0.03·1 votes· comments
You need a Steem account to join the discussion
Sign up now