A Fisher-Rao gradient flow for entropy-regularised Markov decision processes in Polish spaces

Open in new window