Compresses observations into a latent with an autoencoder and learns a recurrent model of how that latent evolves. A controller can then be trained entirely inside it.
supersedescorrects · extends
classifiesspecializes · part-of
substitutes forapproximates · alternative-to
depends onrequires · validates
Colour is the family; a dashed line is the second member of it.
Drag to pan · scroll to zoom · click a node to open it
This node
specializesis a specific case of Model-Based RLthe learned model is a latent-space simulator
requiresdoes not work without Variational Autoencoderthe latent its dynamics run in comes from one
Referenced by
extendsDreamer adds capability to thistrains the policy inside the model, not the environment
extendsGenie adds capability to thisinfers latent actions from video, so none need labelling
correctsMuZero fixes a defect in thispredicting observations is capacity planning cannot use