---
OA_type: closed access
_id: '20667'
abstract:
- lang: eng
  text: We explore the problem of mean estimation for a high-dimensional binary symmetric
    Gaussian mixture model, where the label (sign) follows a time-inhomogeneous Markov
    chain. We propose a spectral estimator based on a partition of a subset of the
    samples to blocks. We develop a computationally efficient algorithm to find the
    optimal blocks, and derive minimax lower bounds on the estimation loss of any
    estimator, which establish the effectiveness of our proposed estimator. The resulting
    minimax rate illuminates the interplay between the sample size, dimension, signal
    strength, and the memory on the loss.
acknowledgement: The research of A.K. and N.W. was supported by the Israel Science
  Foundation (ISF), grant no. 1782/22.
article_processing_charge: No
author:
- first_name: Abd
  full_name: El Latif Kadry, Abd
  last_name: El Latif Kadry
- first_name: Yihan
  full_name: Zhang, Yihan
  id: 2ce5da42-b2ea-11eb-bba5-9f264e9d002c
  last_name: Zhang
  orcid: 0000-0002-6465-6258
- first_name: Nir
  full_name: Weinberger, Nir
  last_name: Weinberger
citation:
  ama: 'El Latif Kadry A, Zhang Y, Weinberger N. Mean estimation in high-dimensional
    binary timeinhomogeneous Markov Gaussian mixture models. In: <i>2025 IEEE International
    Symposium on Information Theory Proceedings</i>. IEEE; 2025. doi:<a href="https://doi.org/10.1109/ISIT63088.2025.11195426">10.1109/ISIT63088.2025.11195426</a>'
  apa: 'El Latif Kadry, A., Zhang, Y., &#38; Weinberger, N. (2025). Mean estimation
    in high-dimensional binary timeinhomogeneous Markov Gaussian mixture models. In
    <i>2025 IEEE International Symposium on Information Theory Proceedings</i>. Ann
    Arbor, MI, United States: IEEE. <a href="https://doi.org/10.1109/ISIT63088.2025.11195426">https://doi.org/10.1109/ISIT63088.2025.11195426</a>'
  chicago: El Latif Kadry, Abd, Yihan Zhang, and Nir Weinberger. “Mean Estimation
    in High-Dimensional Binary Timeinhomogeneous Markov Gaussian Mixture Models.”
    In <i>2025 IEEE International Symposium on Information Theory Proceedings</i>.
    IEEE, 2025. <a href="https://doi.org/10.1109/ISIT63088.2025.11195426">https://doi.org/10.1109/ISIT63088.2025.11195426</a>.
  ieee: A. El Latif Kadry, Y. Zhang, and N. Weinberger, “Mean estimation in high-dimensional
    binary timeinhomogeneous Markov Gaussian mixture models,” in <i>2025 IEEE International
    Symposium on Information Theory Proceedings</i>, Ann Arbor, MI, United States,
    2025.
  ista: 'El Latif Kadry A, Zhang Y, Weinberger N. 2025. Mean estimation in high-dimensional
    binary timeinhomogeneous Markov Gaussian mixture models. 2025 IEEE International
    Symposium on Information Theory Proceedings. ISIT: International Symposium on
    Information Theory.'
  mla: El Latif Kadry, Abd, et al. “Mean Estimation in High-Dimensional Binary Timeinhomogeneous
    Markov Gaussian Mixture Models.” <i>2025 IEEE International Symposium on Information
    Theory Proceedings</i>, IEEE, 2025, doi:<a href="https://doi.org/10.1109/ISIT63088.2025.11195426">10.1109/ISIT63088.2025.11195426</a>.
  short: A. El Latif Kadry, Y. Zhang, N. Weinberger, in:, 2025 IEEE International
    Symposium on Information Theory Proceedings, IEEE, 2025.
conference:
  end_date: 2025-06-27
  location: Ann Arbor, MI, United States
  name: 'ISIT: International Symposium on Information Theory'
  start_date: 2025-06-22
date_created: 2025-11-23T23:01:39Z
date_published: 2025-10-20T00:00:00Z
date_updated: 2025-11-24T08:53:34Z
day: '20'
department:
- _id: MaMo
doi: 10.1109/ISIT63088.2025.11195426
language:
- iso: eng
month: '10'
oa_version: None
publication: 2025 IEEE International Symposium on Information Theory Proceedings
publication_identifier:
  isbn:
  - '9798331543990'
  issn:
  - 2157-8095
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Mean estimation in high-dimensional binary timeinhomogeneous Markov Gaussian
  mixture models
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
year: '2025'
...