Abstract: Techniques are disclosed that enable training a goal-conditioned policy based on multiple data sets, where each of the data sets describes a robot task in a different way. For example, the multiple data sets can include: a goal image data set, where the task is captured in the goal image; a natural language instruction data set, where the task is described in the natural language instruction; a task ID data set, where the task is described by the task ID, etc. In various implementations, each of the multiple data sets has a corresponding encoder, where the encoders are trained to generate a shared latent space representation of the corresponding task description. Additional or alternative techniques are disclosed that enable control of a robot using a goal-conditioned policy network. For example, the robot can be controlled, using the goal-conditioned policy network, based on free-form natural language input describing robot task(s).

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling an agent. One of the methods includes controlling the agent using a policy neural network that processes a policy input that includes (i) a current observation, (ii) a goal observation, and (iii) a selected latent plan to generate a current action output that defines an action to be performed in response to the current observation.