Metraipolis › archive › dc_291f76d650f1

PGFS++: Molecular Property Improvement under Synthesis and Diversity Constraints — arXiv:2608.19121

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.19121
License (as archived)
CC0-1.0
Archived
2026-10-18T13:45:35.535Z
SHA-256 of the archived text
1b337dd5e8e5414f340b73a3edae41708340448b5335488be74b796904499a38
Stable URL of this page
https://metraipolis.com/documento/dc_291f76d650f1

How to cite

PGFS++: Molecular Property Improvement under Synthesis and Diversity Constraints — arXiv:2608.19121 — arXiv (en), https://arxiv.org/abs/2608.19121 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_291f76d650f1 (Metraipolis archive dc_291f76d650f1, sha256 1b337dd5e8e5414f340b73a3edae41708340448b5335488be74b796904499a38)
BibTeX
@misc{dc291f76d650f1,
  title = {PGFS++: Molecular Property Improvement under Synthesis and Diversity Constraints — arXiv:2608.19121},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.19121},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_291f76d650f1; sha256 1b337dd5e8e5414f340b73a3edae41708340448b5335488be74b796904499a38},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19121. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Improving molecular properties, such as drug-likeness or binding affinity, is a recurring task in early-stage drug discovery. However, molecules optimized in an unconstrained chemical space have limited practical value if they cannot be synthesized. Policy Gradient for Forward Synthesis (PGFS) is a synthesis-aware reinforcement learning method for molecular improvement, but its use of reactant embedding prediction makes reactant selection indirect, which, as we show, limits learning effectiveness. We first develop PGFS+, in which reaction templates and second reactants are represented by trainable embedding lookup tables. Combined with a more effective scoring function and RL algorithm, PGFS+ significantly improves the desired property. However, it exposes a reward-hacking failure mode: a powerful reactant search can map diverse input molecules to the same high-reward magnet molecule, improving the reward while collapsing the output diversity. We therefore introduce PGFS++, a synthesis-aware reinforcement learning framework for input-specific molecular improvement. Given an input molecule, PGFS++ treats it as the start of a forward-synthesis trajectory, applies learned reaction templates with compatible in-stock building blocks, and produces a molecule with improved target properties, an explicit synthesis route, and structural similarity to the input. Experiments on molecular improvement tasks show that PGFS++ improves target properties while preserving high output diversity.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.19121 — 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 2026-10-18T13:41:18.002Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).