Distributed Reinforcement Learning

Did not do enough of this at Dyna.

Gabriel Barth-Maron was a pioneer of this.

Some papers:

• Distributed Prioritized Experience Replay
• D4PG