Using temporal neighborhoods to adapt function approximators in reinforcement learning

To avoid the curse of dimensionality, function approximators areused in reinforcement learning to learn value functions for individualstates. In order to make better use of computational resources(basis functions) many researchers are investigating ways to adaptthe basis functions during the learning process so that they betterfit the value-function landscape. Here we introduce temporal neighborhoods as small groups of states that experience frequent intragrouptransitions during...