Abstract
Archive-based exploration methods such as Go-Explore select which visited state to return to using visitation rarity, while frontier methods return to the boundary of the unknown; neither addresses whether the unexplored regions behind a boundary are actually enterable. Exploration is not just about finding rewards; it’s about collecting a structurally complete experience for downstream learning and planning. We introduce TopoExplore, which augments Go-Explore cell selection with a periodic topological pass: enclosed unexplored regions (voids) of the visited-set occupancy grid are detected by flood fill (the H1 classes of its cubical complex), and a decaying selection bonus is placed only on their strict entrances (gap or door cells), ensuring that sealed regions are never targeted and entered regions retire.
In a controlled 18-environment MiniGrid suite (15 seeds, frozen hyperparameters), TopoExplore achieves a 1.52x geometric-mean speedup in median steps-to-first-entry over its exact Go-Explore ablation, compared to 1.37x for a frontier baseline; frontier exploration degrades when sealed decoy structures appear (0.83-1.48x on decoy environments vs. 1.65-2.11x for TopoExplore), while TopoExplore shows its largest win on hard multi-interaction doors (10.9x).
We report an honest negative on Montezuma's Revenge - without wall knowledge, unreachable occupancy artifacts capture the bonus and performance degrades as it grows, isolating the wall-aware entrance test as the load-bearing component - and a preliminary positive on HM3D scanned buildings, where the speedup over Go-Explore tracks scene difficulty (r=0.69) even as frontier selection dominates blanket coverage.
The evidence supports a deliberately scoped claim: topology-aware selection pays off where enclosed structure must be discriminated, and remains competitive at open coverage, where frontier methods are strongest, despite not being tuned for that regime.
Blogger's Review: TopoExplore significantly enhances exploration efficiency by introducing a topology-aware selection mechanism, especially excelling in complex environments. This method not only addresses the limitations of traditional approaches but also provides a more structured experience for subsequent learning and planning, showcasing broad application potential. Future developments in the exploration field are worth watching.