In continuous learning settings stochastic stable policies are often necessary to ensure that agents continuously adapt to dynamic environments. The choice of the decentralised learning system and the employed policy ...