---
res:
  bibo_abstract:
  - We study differentially private model training with stochastic gradient descent
    under learning rate scheduling and correlated noise. Although correlated noise,
    in particular via matrix factorizations, has been shown to improve accuracy, prior
    theoretical work focused primarily on the prefix-sum workload. That workload assumes
    a constant learning rate, whereas in practice learning rate schedules are widely
    used to accelerate training and improve convergence. We close this gap by deriving
    general upper and lower bounds for a broad class of learning rate schedules in
    both single- and multi-epoch settings. Building on these results, we propose a
    learning-rate-aware factorization that achieves improvements over prefix-sum factorizations
    under both MaxSE and MeanSE error metrics. Our theoretical analysis yields memory-efficient
    constructions suitable for practical deployment, and experiments on CIFAR-10 and
    IMDB datasets confirm that schedule-aware factorizations improve accuracy in private
    training.@eng
  bibo_authorlist:
  - foaf_Person:
      foaf_givenName: Nikita
      foaf_name: Kalinin, Nikita
      foaf_surname: Kalinin
      foaf_workInfoHomepage: http://www.librecat.org/personId=4b14526e-14d2-11ed-ba64-c14c9553d137
  - foaf_Person:
      foaf_givenName: Joel D
      foaf_name: Andersson, Joel D
      foaf_surname: Andersson
      foaf_workInfoHomepage: http://www.librecat.org/personId=4a893819-d954-11f0-89b1-e360bad9ccc5
  bibo_doi: 10.4230/LIPIcs.FORC.2026.2
  bibo_volume: 368
  dct_date: 2026^xs_gYear
  dct_isPartOf:
  - http://id.crossref.org/issn/1868-8969
  - http://id.crossref.org/issn/9783959774192
  dct_language: eng
  dct_publisher: Schloss Dagstuhl - Leibniz-Zentrum für Informatik@
  dct_subject:
  - differential privacy
  - machine learning
  - matrix factorization
  dct_title: Learning rate scheduling with matrix factorization for private training@
...