Planner-Admissible Graph-PDE Value Extensions for Sparse Goal-Conditioned Planning
About
Sparse goal-conditioned planning with few cost-to-go labels can be viewed as a graph-PDE Dirichlet extension problem: extend sparse labels on a goal-dependent boundary to unlabelled graph vertices so that greedy rollouts reach the goal. We study which graph value extensions are planner-admissible under the operational argmin-Q planner. Our main result is a local action-gap certificate: if the surrogate value error along the rollout stays below half the true action gap, then the greedy rollout reaches the goal. Absolutely Minimal Lipschitz Extension (AMLE), the p=infinity endpoint of the graph p-Laplacian family, instantiates this certificate through a comparison-principle fill-distance bound. Harmonic extension, by contrast, can mis-rank local actions because its values reflect boundary hitting probabilities rather than shortest-path greedy order. On 120 AntMaze layout-derived graph configurations, harmonic extension achieves 0.584 aggregate rollout success, while AMLE reaches 0.970. Finite high-p methods also enter a high-success regime, with success 0.903 for p=4, 0.973 for p=8, and 0.982 for a fixed-budget p=16 solver, though the p=16 row is not used as a converged endpoint ranking due to incomplete solver certification. Mechanism audits show that many rollout decisions occur in AMLE-compatible but harmonic-incompatible local geometry, and that AMLE corrects most harmonic inversions on the rollout-weighted decision scope.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Navigation | AntMaze v4 | Success Rate99.3 | 10 | |
| p-family ordering audit | 120-cell main rollout grid | Success Rate98.2 | 5 | |
| Maze Navigation | AntMaze r=8 | Success Rate97.1 | 3 | |
| Maze Navigation | AntMaze r=12 | Success Rate94.6 | 3 | |
| Navigation | AntMaze 120-cell grid paired configurations | Success Rate (Mean)97 | 2 |