Generalized Planning on Floortile
62ScaleHeuristically regularized Q-value policy (Ω^Heu.)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Heuristically regularized Q-value policy (Ω^Heu.)Architecture=R-GNN2026.03 | 62 | 45.6 | |
| Heuristically regularized Q-value policy (Ω^Heu.)Architecture=OAE2026.03 | 62 | 45.9 | |
| Explicitly regularized Q-value policy (Ω^Exp.)Architecture=R-GNN2026.03 | 60 | 44.5 | |
| Explicitly regularized Q-value policy (Ω^Exp.)Architecture=OAE2026.03 | 59 | 42 | |
| Heuristically regularized Q-value policy (Ω^Heu.)Architecture=OE2026.03 | 56 | 43.2 | |
| Explicitly regularized Q-value policy (Ω^Exp.)Architecture=OE2026.03 | 50 | 28.8 | |
| State-value policy (V)Architecture=R-GNN2026.03 | 8 | 0 | |
| Vanilla Q-value policy (Q)Architecture=R-GNN2026.03 | 8 | 0 | |
| State-value policy (V)Architecture=OE2026.03 | 8 | 0 | |
| Vanilla Q-value policy (Q)Architecture=OE2026.03 | 8 | 0 | |
| State-value policy (V)Architecture=OAE2026.03 | 8 | 0 | |
| Vanilla Q-value policy (Q)Architecture=OAE2026.03 | 8 | 0 |