Metraipolis › archive › dc_e1f97f49cb6d

Preference-Conditioned Group-Relative Policy Optimization for Terminally Constrained Multi-Objective Control — arXiv:2610.41884

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41884
License (as archived)
CC0-1.0
Archived
2026-11-03T08:32:42.987Z
SHA-256 of the archived text
eb18e2120fc2629c8124a33f92ea188f8b636d391646da4d2116079007929f33
Stable URL of this page
https://metraipolis.com/documento/dc_e1f97f49cb6d

How to cite

Preference-Conditioned Group-Relative Policy Optimization for Terminally Constrained Multi-Objective Control — arXiv:2610.41884 — arXiv (en), https://arxiv.org/abs/2610.41884 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1f97f49cb6d (Metraipolis archive dc_e1f97f49cb6d, sha256 eb18e2120fc2629c8124a33f92ea188f8b636d391646da4d2116079007929f33)
BibTeX
@misc{dce1f97f49cb6d,
  title = {Preference-Conditioned Group-Relative Policy Optimization for Terminally Constrained Multi-Objective Control — arXiv:2610.41884},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41884},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1f97f49cb6d; sha256 eb18e2120fc2629c8124a33f92ea188f8b636d391646da4d2116079007929f33},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41884. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-objective optimal control is commonly addressed by reinforcement learning with fixed scalarization weights, requiring a separate training run and constraint enforcement for each trade-off. This letter extends critic-free group-relative policy optimization to terminally constrained multi-objective control: a single preference-conditioned policy is trained over random preferences and returns a feasible approximation of the Pareto front in one run. Three structural results are established. Group normalization after scalarization preserves the prescribed preference, whereas per-objective normalization induces policy-dependent distortion; with pairwise distinct normalization scales, only single-objective preferences remain unchanged. A positive floor prevents the terminal-constraint multiplier from vanishing and renders the penalty exact when the floor exceeds a geometric multiplier. Deterministic terminal violation is bounded by expected stochastic violation plus a term depending on terminal-state curvature and action variance, so stochastic feasibility alone does not certify deterministic execution. On a heavy-duty range-extender powertrain with six stage costs and a terminal state-of-charge band, a representative training run places all 32 preference-grid solutions in band on six cycles inside and outside the training pool. Averaged over ten runs, the method yields more in-band solutions under deterministic execution than a critic-based conditioned baseline with the same terminal-constraint handling.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv eess.SY, nuovi lavori del giorno).
Categorie arXiv: Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2610.41884 — 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 2026-11-03T08:15:24.235Z dalla pagina https://arxiv.org/list/eess.SY/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).