Guardado en:
Detalles Bibliográficos
Autores principales: Suman, Uttam, Mamajiwala, Mariya, Saxena, Mukul, Tyagi, Ankit, Roy, Debasish
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2410.14270
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913712123674624
author Suman, Uttam
Mamajiwala, Mariya
Saxena, Mukul
Tyagi, Ankit
Roy, Debasish
author_facet Suman, Uttam
Mamajiwala, Mariya
Saxena, Mukul
Tyagi, Ankit
Roy, Debasish
contents Our proposal is on a new stochastic optimizer for non-convex and possibly non-smooth objective functions typically defined over large dimensional design spaces. Towards this, we have tried to bridge noise-assisted global search and faster local convergence, the latter being the characteristic feature of a Newton-like search. Our specific scheme -- acronymed FINDER (Filtering Informed Newton-like and Derivative-free Evolutionary Recursion), exploits the nonlinear stochastic filtering equations to arrive at a derivative-free update that has resemblance with the Newton search employing the inverse Hessian of the objective function. Following certain simplifications of the update to enable a linear scaling with dimension and a few other enhancements, we apply FINDER to a range of problems, starting with some IEEE benchmark objective functions to a couple of archetypal data-driven problems in deep networks to certain cases of physics-informed deep networks. The performance of the new method vis-á-vis the well-known Adam and a few others bears evidence to its promise and potentialities for large dimensional optimization problems of practical interest.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14270
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FINDER: Stochastic Mirroring of Noisy Quasi-Newton Search and Deep Network Training
Suman, Uttam
Mamajiwala, Mariya
Saxena, Mukul
Tyagi, Ankit
Roy, Debasish
Machine Learning
Our proposal is on a new stochastic optimizer for non-convex and possibly non-smooth objective functions typically defined over large dimensional design spaces. Towards this, we have tried to bridge noise-assisted global search and faster local convergence, the latter being the characteristic feature of a Newton-like search. Our specific scheme -- acronymed FINDER (Filtering Informed Newton-like and Derivative-free Evolutionary Recursion), exploits the nonlinear stochastic filtering equations to arrive at a derivative-free update that has resemblance with the Newton search employing the inverse Hessian of the objective function. Following certain simplifications of the update to enable a linear scaling with dimension and a few other enhancements, we apply FINDER to a range of problems, starting with some IEEE benchmark objective functions to a couple of archetypal data-driven problems in deep networks to certain cases of physics-informed deep networks. The performance of the new method vis-á-vis the well-known Adam and a few others bears evidence to its promise and potentialities for large dimensional optimization problems of practical interest.
title FINDER: Stochastic Mirroring of Noisy Quasi-Newton Search and Deep Network Training
topic Machine Learning
url https://arxiv.org/abs/2410.14270