Semantic Scholar Open Access 2019 4 sitasi

Astro for Derivative-Based Stochastic Optimization: Algorithm Description & Numerical Experiments

Daniel Vasquez R. Pasupathy S. Shashaani

Abstrak

Adaptive Sampling Trust-Region Optimization (ASTRO) is a class of derivative-based stochastic trust-region algorithms developed to solve stochastic unconstrained optimization problems where the objective function and its gradient are observable only through a noisy oracle or using a large dataset. ASTRO incorporates adaptively sampled function and gradient estimates within a trust-region framework to generate iterates that are guaranteed to converge almost surely to a first-order or a second-order critical point of the objective function. Efficiency in ASTRO stems from two key aspects: (i) adaptive sampling to ensure that the objective function and its gradient are sampled only to the extent needed, so that small sample sizes result when iterates are far from a critical point and large sample sizes result when iterates are near a critical point; and (ii) quasi-Newton Hessian updates using BFGS. We describe ASTRO in detail, give a sense of its theoretical guarantees, and report extensive numerical results.

Topik & Kata Kunci

Penulis (3)

D

Daniel Vasquez

R

R. Pasupathy

S

S. Shashaani

Format Sitasi

Vasquez, D., Pasupathy, R., Shashaani, S. (2019). Astro for Derivative-Based Stochastic Optimization: Algorithm Description & Numerical Experiments. https://doi.org/10.1109/WSC40007.2019.9004904

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →
Lihat di Sumber doi.org/10.1109/WSC40007.2019.9004904
Informasi Jurnal
Tahun Terbit
2019
Bahasa
en
Total Sitasi
Sumber Database
Semantic Scholar
DOI
10.1109/WSC40007.2019.9004904
Akses
Open Access ✓