Beyond Imitation: Self-Improving Robot Policies via Off-Policy Q-Planning now lives at q-planning.github.io.
You should be redirected automatically. If not, follow the link above.