The package contains a finite horizon MDP solver for discrete problems. This algorithm is a modified version of infinite horizon value iteration.
To install FiniteHorizonValueIteration, run the following command:
using Pkg
Pkg.add("FiniteHorizonValueIteration")using POMDPs
using FiniteHorizonValueIteration
mdp = MyMDP() # initialize MDP
# initialize the solver
solver = FiniteHorizonSolver(verbose=false)
# run the solver
policy = solve(solver, mdp)The policy is stored to FiniteHorizonPolicy.
Example problems are defined in test/instances/..., examples are used in corresponding test files. Results are validated against the value iteration executed on all epochs simultaneously.