Reinforcement learning requires information about states, actions, and outcomes as the basis for learning. For many applications,\nit can be difficult to construct a representative model of the environment, either due to lack of required information or because\nof that the model�s state space may become too large to allow a solution in a reasonable amount of time, using the experience of\nprior actions. An environment consisting solely of the occurrence or nonoccurrence of specific events attributable to a human\nactor may appear to lack the necessary structure for the positioning of responding agents in time and space using reinforcement\nlearning. Digital pheromones can be used to synthetically augment such an environment with event sequence information to create\namore persistent and measurable imprint on the environment that supports reinforcement learning.We implemented this method\nand combined it with the ability of agents to learn from actions not taken, a concept known as fictive learning. This approach was\ntested against the historical sequence of Somali maritime pirate attacks from2005 to mid-2012, enabling a set of autonomous agents\nrepresenting naval vessels to successfully respond to an average of 333 of the 899 pirate attacks, outperforming the historical record\nof 139 successes
Loading....