Deep Multiagent Reinforcement Learning for Partially Observable Parameterized Environments
As software and hardware agents begin to perform tasks of genuine interest, they will be faced with environments too complex for humans to predetermine the correct actions to take. Three characteristics shared by many complex…