A Deep RL method where many workers learn in parallel without waiting.
A3C is like a video game speedrun club. Several players try routes at once. Then they shout tips across the couch.
It helped train game AIs faster. It also helped control things inside simulators.
Actor-Critic
A3C is the async parallel version of Actor-Critic.
Deep RL
A3C was an early landmark algorithm in Deep RL.
Policy Gradient
A3C uses Policy Gradient to update the actor's action policy.
On-Policy Learning
A3C usually samples with the current policy and learns right away.