Optimizing Latent Goal by Learning from Trajectory Preference

Open in new window