Accompanying code for "Learning and Planning in Average-Reward Markov Decision Processes" - View it on GitHub