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.