In mhahsler/pomdp: Infrastructure for Partially Observable Markov Decision Processes (POMDP)

pkg <- 'pomdp'

source("https://raw.githubusercontent.com/mhahsler/pkg_helpers/main/pkg_helpers.R")
pkg_title(pkg)

Introduction

A partially observable Markov decision process (POMDP) models an agent decision process where the agent cannot directly observe the environment's state, but has to rely on observations. The goal is to find an optimal policy to guide the agent's actions.

The pomdp package provides the infrastructure to define and analyze the solutions of optimal control problems formulated as Partially Observable Markov Decision Processes (POMDP). The package uses the solvers from pomdp-solve (Cassandra, 2015) available in the companion R package pomdpSolve to solve POMDPs using a variety of exact and approximate algorithms.

The package provides fast functions (using C++, sparse matrix representation, and parallelization with foreach) to perform experiments (sample from the belief space, simulate trajectories, belief update, calculate the regret of a policy). The package also interfaces to the following algorithms:

• Exact value iteration
• Enumeration algorithm [@Sondik1971; @Monahan1982].
• Two pass algorithm [@Sondik1971].
• Witness algorithm [@Littman1995].

• Incremental pruning algorithm [@Zhang1996; @Cassandra1997].

• Approximate value iteration

• Finite grid algorithm [@Cassandra2015], a variation of point-based value iteration to solve larger POMDPs (PBVI; see [@Pineau2003] without dynamic belief set expansion.
• SARSOP [@Kurniawati2008], point-based algorithm that approximates optimally reachable belief spaces for infinite-horizon problems (via package sarsop).

If you are new to POMDPs then start with the POMDP Tutorial.

pkg_citation(pkg, 1)
pkg_install(pkg)

Usage

Solving the simple infinite-horizon Tiger problem.

library("pomdp")
data("Tiger")
Tiger

sol <- solve_POMDP(model = Tiger)
sol

Display the value function.

plot_value_function(sol, ylim = c(0, 20))

Display the policy graph.

plot_policy_graph(sol)

Acknowledgments

Development of this package was supported in part by National Institute of Standards and Technology (NIST) under grant number 60NANB17D180.

References

mhahsler/pomdp documentation built on Dec. 8, 2024, 4:26 a.m.