2025/09/02 by David Beers, Beers, David, Gillian Grindstaff +1
Mathematics · #55M99 (primary) 51F99 (Secondary) #Algebraic Topology (math.AT) #FOS: Mathematics #Metric Geometry (math.MG) #math.AT #math.MG #msc:51F99 #msc:55M99
paper · pdf · doi:10.48550/arxiv.2509.02755
v2: Reworked the proof of the main theorem (now Theorem 21) after identifying a mistake in the proof of the previous Proposition 21 (the previous Equation (4) is in general false). Added several further results (see especially Corollary 22 and Theorem 23). 20 pages, 4 figures
arxiv created 2026/07/29 · arxiv updated 2026/07/30
Merge trees are a topological descriptor of a filtered space that enriches the degree zero barcode with its merge structure. The space of merge trees comes equipped with an interleaving distance dI, which prompts the natural question: is the interleaving distance between two merge trees equal to the bottleneck distance between their corresponding barcodes? As the map from merge trees to barcodes is not injective, the answer as posed is no, but (as conjectured in Gasparovic et al.) we prove that it is true for the intrinsic metrics \widehatdI and \widehatdB realized by infinitesimal path length in merge tree space. This implies that they have the same induced length space, and that in cases where a path is known, the bottleneck distance (which can be computed quickly) can be substituted for the interleaving distance (in general, NP-hard to approximate).