Accenture partnered with San Francisco based AI company Pathmind to investigate the potential of new reinforcement learning (RL) opportunities in simulation.
The results obtained were extremely good. The method produced a waiting time more than 4x shorter than the Nearest Agent heuristic.
In this blog post, Agustin Albinati summarizes the model, introduces the three key considerations when defining the neural net, and presents the results of his team's investigations. Linked at the end of the blog post is a step by step how-to with Pathmind. Read on!