PAPERS-03-001 · Paper — The Lineage — How Machines Learned to Act

The paper

Before this paper

The key idea

How it works

See it — the deep RL lineage

flowchart LR
  A["DQN<br/>learn to play"] --> B["A3C / PPO<br/>stable policies"]
  B --> C["AlphaGo<br/>search + learning"]
  C --> D["AlphaZero<br/>pure self-play"]
  D --> E["MuZero<br/>learned model"]

What they showed

What it changed

Next: Deep Q-Networks (DQN).

The one-line takeaway

Read it / meet it