Metraipolis › archive › dc_09353cd51d3f

Beyond IPS: Reliable Counterfactual Evaluation in Multi-Stage Ad Systems without Logged Propensities — arXiv:2610.40112

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.40112
License (as archived)
CC0-1.0
Archived
2026-11-01T02:40:01.249Z
SHA-256 of the archived text
919c416b6c8a16f5c5cda2ad9e431240b3a9f81343685190c94040684ef80e3e
Stable URL of this page
https://metraipolis.com/documento/dc_09353cd51d3f

How to cite

Beyond IPS: Reliable Counterfactual Evaluation in Multi-Stage Ad Systems without Logged Propensities — arXiv:2610.40112 — arXiv (en), https://arxiv.org/abs/2610.40112 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09353cd51d3f (Metraipolis archive dc_09353cd51d3f, sha256 919c416b6c8a16f5c5cda2ad9e431240b3a9f81343685190c94040684ef80e3e)
BibTeX
@misc{dc09353cd51d3f,
  title = {Beyond IPS: Reliable Counterfactual Evaluation in Multi-Stage Ad Systems without Logged Propensities — arXiv:2610.40112},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40112},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09353cd51d3f; sha256 919c416b6c8a16f5c5cda2ad9e431240b3a9f81343685190c94040684ef80e3e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40112. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Offline evaluation of ad-ranking policies in multi-stage delivery systems is challenging because action propensities are often unavailable, limiting inverse propensity scoring (IPS). Reward-model-based estimators such as the Direct Method (DM) avoid propensities but can be biased when candidate policies induce impression distributions that differ from the logged data. We formalize DM-based counterfactual replay and propose mixture-weighted reward modeling, which reweights logged data toward a mixture of candidate-policy distributions using density ratios from a multiclass domain classifier. We bound each candidate's value-estimation error by overlap, cold-start, and replay terms, and derive an asymptotic rule that uses the cold-start mass to decide whether to trust offline selection or route candidates to online A/B testing. On an ads-delivery simulator and on production traffic from a large-scale CTR ranking system, the mixture-weighted reward model improves policy selection over its unweighted counterpart in both settings, with the largest gains at low overlap in simulation. Combined with a doubly robust correction, it attains the highest Top-1 accuracy in production and at low and medium overlap in simulation.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.IR, nuovi lavori del giorno).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2610.40112 — 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 2026-11-01T02:16:28.230Z dalla pagina https://arxiv.org/list/cs.IR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).