Model-free safe policy learning via hard action barrier functions.

Agustin Castellano, Juan Andrs Bazerque, Enrique Mallada

Browse the full CISS paper archive.