Metraipolis › archive › dc_eefe84524e77

Beyond One Epoch: Uncertainty-Weighted Sensitivity Regularization for Recommendation Models — arXiv:2609.34083

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34083
License (as archived)
CC0-1.0
Archived
2026-10-07T22:37:59.180Z
SHA-256 of the archived text
9fa2e1230286974b480f3965a8b71876d6b25abe729baa1cefe6bf86c2c525f3
Stable URL of this page
https://metraipolis.com/documento/dc_eefe84524e77

How to cite

Beyond One Epoch: Uncertainty-Weighted Sensitivity Regularization for Recommendation Models — arXiv:2609.34083 — arXiv (en), https://arxiv.org/abs/2609.34083 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eefe84524e77 (Metraipolis archive dc_eefe84524e77, sha256 9fa2e1230286974b480f3965a8b71876d6b25abe729baa1cefe6bf86c2c525f3)
BibTeX
@misc{dceefe84524e77,
  title = {Beyond One Epoch: Uncertainty-Weighted Sensitivity Regularization for Recommendation Models — arXiv:2609.34083},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34083},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eefe84524e77; sha256 9fa2e1230286974b480f3965a8b71876d6b25abe729baa1cefe6bf86c2c525f3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34083. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recommendation models with sparse embeddings and a shared consumer often exhibit the one-epoch phenomenon: a second epoch lowers training loss while sharply degrading generalization. We present a view based on the violation of the prequential principle. On the first epoch, an example's label has not affected the embedding rows used to score it. On later epochs, those rows contain a displacement induced by the labels earlier update. This creates an incentive for the shared consumer to exploit this displacement in subsequent epochs, which fails to generalize. We call this self-influence asymmetry. Using an exact scalar model and local influence analysis, we connect this mismatch to the uncertainty in the embeddings and the consumers incentive to exploit it in subsequent epochs. We verify this hypothesis using an embedding-consumer-update interventions in deep recommendation models and propose uncertainty-weighted sensitivity regularization (UWSR) which counteracts this mismatch by augmenting the loss function to penalize the consumer for relying on uncertain embeddings. Unlike existing remedies, UWSR preserves the learned embeddings and across three benchmarks, four-epoch UWSR reduces test cross-entropy by 1.38%-6.78% and improves AUC by 0.0058-0.0231 relative to one-epoch training.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.34083 — 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 2026-10-07T22:14:28.106Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).