non-markovian
Approximate Numerics in Applied Transition Environments: finite MDPs and history-dependent decision problems in Python