2 Hits in 4.2 sec

Artifact for "Teaching Stratego to Play Ball : Optimal Synthesis for Continuous Space MDPs" [article]

Peter Gjøl Jensen
2019 Zenodo  
A zip-file containing the artifact, models and scripts for reproducing the results of the paper "Teaching Stratego to Play Ball : Optimal Synthesis for Continuous Space MDPs" accepted at ATVA'19.  ...  The Artifact Evaluation Package is configured to match the Virtual Machine provided: 10.5281/zenodo.2759473 A guide of the experiments can be found in README.html.  ...  states G Output: A near-optimal, strategy σ for the MDP M under the cost function C for the goal G. 1 Initially let σ(s)(α) = 1 |Act| for any s ∈ R K and any α ∈ Act 2 while Termination criterion is not  ... 
doi:10.5281/zenodo.3268381 fatcat:y2w224amojfg5m6k7ivinjwgge

An Anytime Algorithm for Reachability on Uncountable MDP [article]

Kush Grover, Jan Křetínský, Tobias Meggendorfer, Maximilian Weininger
2020 arXiv   pre-print
We provide an algorithm for reachability on Markov decision processes with uncountable state and action spaces, which, under mild assumptions, approximates the optimal value to any desired precision.  ...  Moreover, it simultaneously is the first algorithm able to utilize learning approaches without sacrificing guarantees and it further allows for combination with existing heuristics.  ...  Teaching stratego to play ball: Optimal synthesis for continuous space mdps.  ... 
arXiv:2008.04824v1 fatcat:54e5csqhojhdpc6brhwero4plq