Reusing previously trained models is critical in deep reinforcement learning to speed up training of new agents. However, it is unclear how to acquire new skills when objectives and constraints are in conflict with previously learned skills. In this presentation, inspired by neuroscience research, we propose a novel approach using inhibitory networks to allow separate and adaptive state value evaluations, as well as distinct automatic entropy tuning.

There are still a number of internships and full-time position available. Please visit www.lockheedmartinjobs.com.