Metraipolis › archive › dc_4e3d2bd71e44

Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation — arXiv:2610.08743

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.08743
License (as archived)
CC0-1.0
Archived
2026-10-07T11:22:08.920Z
SHA-256 of the archived text
249509eedf2c3a20f42f8283a05e6a4074fe72479cb7d2f9cd9898fe2db7cc90
Stable URL of this page
https://metraipolis.com/documento/dc_4e3d2bd71e44

How to cite

Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation — arXiv:2610.08743 — arXiv (en), https://arxiv.org/abs/2610.08743 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4e3d2bd71e44 (Metraipolis archive dc_4e3d2bd71e44, sha256 249509eedf2c3a20f42f8283a05e6a4074fe72479cb7d2f9cd9898fe2db7cc90)
BibTeX
@misc{dc4e3d2bd71e44,
  title = {Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation — arXiv:2610.08743},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.08743},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4e3d2bd71e44; sha256 249509eedf2c3a20f42f8283a05e6a4074fe72479cb7d2f9cd9898fe2db7cc90},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08743. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sequential recommenders typically use a fixed slate size even though the number of useful alternatives changes within a session. We propose Reinforcement Learning with Calibrated Pruning (RLCP), which adapts the retained action set using critic scores and an online threshold. The threshold is updated from binary feedback indicating whether the set contains an action in a proxy target. We prove a deterministic bound on the observed proxy miss rate along adaptive trajectories. To quantify the effect of pruning on reward, we derive an exact decomposition of value loss into filtering and selection losses. Under explicit proxy and critic approximation conditions, this decomposition yields a finite session reward bound that also accounts for imperfect selection and set truncation, without requiring the learning parameters to converge. Experiments on KuaiRand-Pure and MovieLens 1M compare two RLCP implementations with four RL baselines. In each of the 19 configurations, at least one RLCP variant achieves the highest catalog diversity, reaching $1.11\times$ to $5.21\times$ that of the strongest baseline, with competitive session depth and no larger retained sets.

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).
Pagina del lavoro: https://arxiv.org/abs/2610.08743 — 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 2026-10-07T11:15:24.863Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).