Metraipolis › archive › dc_810d3039599e

Reward-Weighted Preference Optimization for Selective Combinatorial Optimization Problems — arXiv:2610.31239

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31239
License (as archived)
CC0-1.0
Archived
2026-10-25T20:53:21.217Z
SHA-256 of the archived text
406b55e84d96eb1776fa9878cf7bb96012eca56faba8ad3594e266bb525383f2
Stable URL of this page
https://metraipolis.com/documento/dc_810d3039599e

How to cite

Reward-Weighted Preference Optimization for Selective Combinatorial Optimization Problems — arXiv:2610.31239 — arXiv (en), https://arxiv.org/abs/2610.31239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_810d3039599e (Metraipolis archive dc_810d3039599e, sha256 406b55e84d96eb1776fa9878cf7bb96012eca56faba8ad3594e266bb525383f2)
BibTeX
@misc{dc810d3039599e,
  title = {Reward-Weighted Preference Optimization for Selective Combinatorial Optimization Problems — arXiv:2610.31239},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31239},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_810d3039599e; sha256 406b55e84d96eb1776fa9878cf7bb96012eca56faba8ad3594e266bb525383f2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning has emerged as an effective approach for combinatorial optimization, but most existing methods focus on problems such as the Travelling Salesperson Problem (TSP) in which all elements must be served. Selective combinatorial optimization problems (SCOPs), including the Orienteering Problem (OP), Prize-Collecting TSP (PCTSP), Stochastic PCTSP and Knapsack Problem (KP), additionally require selecting which elements to include. This selection creates sampled solutions with substantially different objective values, yet standard preference optimization reduces such comparisons to binary rankings and discards the magnitude of these differences. We introduce Reward-Weighted Preference Optimization (RWPO), a pairwise training objective that incorporates reward-gap information into preference learning. For candidate trajectories sampled from the same instance, RWPO weights each preference loss according to the normalized difference in their rewards, assigning greater training emphasis to comparisons that represent larger improvements in solution quality. We further combine RWPO with scale-specific fine-tuning and reward- and likelihood-weighted tilted sampling to adapt trained policies to larger and shifted instance distributions without relying on local-search post-processing. Experiments on the OP, PCTSP, Stochastic PCTSP, and KP show that RWPO consistently improves over unweighted preference optimization and policy-gradient baselines across a wide range of instance sizes. On the OP, the resulting solver exceeds the mean reward of the Compass metaheuristic at scales from 100 to 2,000 nodes. The results demonstrate that preserving reward-gap information can substantially improve preference-based learning for selective combinatorial optimization and support generalization to large-scale instances.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.31239 — 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 2026-10-25T20:47:56.919Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).