ResearchBenchmarksCollect on Collect Deterministic OriginalFollow9.71Average RewardShortest8.5668.8639.169.457Jul 11, 2017Evaluation ResultsMethodMethodLinksAverage RewardShortest2017.079.71OPNplanning steps=5planning steps=52017.079.3VPNplanning steps=5planning steps=52017.079.29VPNplanning steps=1planning steps=12017.078.94DQN2017.078.66Greedy2017.078.61