Arxiu de l'autor: De Curtò i DíAz

Studio Ghibli

Publicat dins de General | Comentaris tancats a Studio Ghibli

Evolved Policy Gradients (EPG)

Evolved Policy Gradients, an experimental meta-learning technique to let agents rapidly learn to solve novel tasks: https://blog.openai.com/evolved-policy-gradients/

Publicat dins de General | Comentaris tancats a Evolved Policy Gradients (EPG)