f-Policy Gradients: A General Framework for Goal-Conditioned RL using f-Divergences
Posted on