Biblio Discovery — Demo Discovery Service

4 hasil untuk "Smail Addoune"

Menampilkan 1–4 · 1,67 detik · RSS

  1. Artikel Jurnal Teks lengkap

    Tailed-Average SGD: A Polyak-Ruppert modification for optimal-rate constrained optimization

    Abdelouakil Fillali; Smail Addoune; Rebiha Zeghdane

    Iranian Journal of Numerical Analysis and Optimization, Vol. 16 No. 2 · 2026 · Inggris · DOAJ Scopus

    Stochastic Gradient Descent algorithms are a cornerstone of large-scale optimization, yet their application to problems with explicit constraints remains a significant challenge. While methods based on relaxed barrier functions offer a promising alternative that avoids computationally expensive…

  2. Artikel Jurnal Teks lengkap

    Practical early stopping for adaptive barrier SGD: balancing stochastic speed with validation accuracy

    Abdelouakil Fillali; Smail Addoune; Rebiha Zeghdane

    Iranian Journal of Numerical Analysis and Optimization, Vol. 16 No. Issue 4 · 2026 · Inggris · DOAJ Scopus

    Stochastic Gradient Descent (SGD) algorithms based on adaptive barrier functions are highly efficient for large-scale constrained optimization. However, their inherent stochastic nature leads to the critical ”last- iterate problem”, where update noise causes significant oscillations, making…

  3. Artikel Jurnal Teks lengkap

    Adaptive-batch stochastic gradient descent for constrained optimization based on relaxed barrier functions

    Abdelouakil Fillali; Smail Addoune; Rebiha Zeghdane

    Iranian Journal of Numerical Analysis and Optimization, Vol. 16 No. Issue 3 · 2026 · Inggris · DOAJ Scopus

    Stochastic Gradient Descent (SGD) is the cornerstone of large-scale optimization; however, its application to problems with a vast number of constraints remains a significant challenge. Methods based on relaxed logarithmic barrier functions have enabled the use of SGD by sampling constraints, yet…

  4. Artikel Jurnal Teks lengkap

    Robust stochastic gradient descent for linearly constrained problems via adaptive barrier amplification

    Abdelouakil Fillali; Smail Addoune; Rebiha Zeghdane

    Iranian Journal of Numerical Analysis and Optimization, Vol. 16 No. Issue 3 · 2026 · Inggris · DOAJ Scopus

    Methods based on Stochastic Gradient Descent using relaxed barrier functions provide a powerful framework for linearly constrained optimization but often suffer from sensitivity to hyperparameter settings and slow recovery from infeasible regions. Our investigation reveals that the optimal…