1 paper
Matilde Gargiani, Robin Sieber, Efe Balta +2
We consider inexact policy iteration methods for large-scale infinite-horizon discounted MDPs with finite spaces, a variant of policy iteration where the policy evaluation step is…