Novelty producing synaptic plasticity

A learning process with the plasticity property often requires reinforcement signals to guide the process. However, in some tasks (e.g. maze-navigation), it is very difficult to measure the performance of an agent to provide reinforcements, since the position of the goal is not known. This requires finding the correct behavior among a vast number of possible behaviors without having any feedback. In these cases, an exhaustive search may be needed. However, this might not be feasible especially when optimizing artificial neural networks in continuous domains. In this work, we introduce novelty producing synaptic plasticity (NPSP), where we evolve synaptic plasticity rules to produce as many novel behaviors as possible to find the behavior that can solve the problem. We evaluate the NPSP on deceptive maze environments that require the achievement of subgoals. Our results show that the proposed NPSP produces more novel behaviors compared to Random Search and Random Walk.