Metraipolis › archive › dc_1cf4baf67cf2

HexIt: Learning Hexahedral Block Decomposition with Expert Iteration on Combinatorial Maps — arXiv:2611.10433

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.10433
License (as archived)
CC0-1.0
Archived
2026-11-10T18:42:29.610Z
SHA-256 of the archived text
44727ade86f097edbef88b0f68ac96406890910ec62ff7005861c115d10a364e
Stable URL of this page
https://metraipolis.com/documento/dc_1cf4baf67cf2

How to cite

HexIt: Learning Hexahedral Block Decomposition with Expert Iteration on Combinatorial Maps — arXiv:2611.10433 — arXiv (en), https://arxiv.org/abs/2611.10433 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1cf4baf67cf2 (Metraipolis archive dc_1cf4baf67cf2, sha256 44727ade86f097edbef88b0f68ac96406890910ec62ff7005861c115d10a364e)
BibTeX
@misc{dc1cf4baf67cf2,
  title = {HexIt: Learning Hexahedral Block Decomposition with Expert Iteration on Combinatorial Maps — arXiv:2611.10433},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10433},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1cf4baf67cf2; sha256 44727ade86f097edbef88b0f68ac96406890910ec62ff7005861c115d10a364e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10433. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Decomposing a polyhedral solid into a few well-shaped hexahedral blocks is a long-standing open problem in mesh generation. We cast it as a sequential decision problem on a 3-dimensional generalised map: an agent splits edges, faces and volumes with three insert operations until every cell is a hexahedron and the mesh passes a quality threshold, and a policy that convolves over the map's involutions chooses each insert. A successful decomposition takes over a hundred decisions, and a uniformly random policy never finds one. We clone a policy from constructed decompositions and improve it by expert iteration. On held-out chamfered solids that share no boundary topology with the cloning data, the final policy solves an expected $94.4\%$ in $20$ samples, and expert iteration raises the best-of-20 solve rate on large chamfered solids from $0.203$ to $0.355$. At matched compute, PPO, GRPO and VinePPO updates all make the cloned policy worse.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computational Geometry (cs.CG).
Pagina del lavoro: https://arxiv.org/abs/2611.10433 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-10T18:37:37.588Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).