Improving Gradient Methods via Coordinate Transformations: Applications to Quantum Machine Learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bermejo, Pablo, Aizpurua, Borja, Orus, Roman
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929326871543808
author Bermejo, Pablo
Aizpurua, Borja
Orus, Roman
author_facet Bermejo, Pablo
Aizpurua, Borja
Orus, Roman
contents Machine learning algorithms, both in their classical and quantum versions, heavily rely on optimization algorithms based on gradients, such as gradient descent and alike. The overall performance is dependent on the appearance of local minima and barren plateaus, which slow-down calculations and lead to non-optimal solutions. In practice, this results in dramatic computational and energy costs for AI applications. In this paper we introduce a generic strategy to accelerate and improve the overall performance of such methods, allowing to alleviate the effect of barren plateaus and local minima. Our method is based on coordinate transformations, somehow similar to variational rotations, adding extra directions in parameter space that depend on the cost function itself, and which allow to explore the configuration landscape more efficiently. The validity of our method is benchmarked by boosting a number of quantum machine learning algorithms, getting a very significant improvement in their performance.
format Preprint
id arxiv_https___arxiv_org_abs_2304_06768
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improving Gradient Methods via Coordinate Transformations: Applications to Quantum Machine Learning
Bermejo, Pablo
Aizpurua, Borja
Orus, Roman
Quantum Physics
Machine Learning
Machine learning algorithms, both in their classical and quantum versions, heavily rely on optimization algorithms based on gradients, such as gradient descent and alike. The overall performance is dependent on the appearance of local minima and barren plateaus, which slow-down calculations and lead to non-optimal solutions. In practice, this results in dramatic computational and energy costs for AI applications. In this paper we introduce a generic strategy to accelerate and improve the overall performance of such methods, allowing to alleviate the effect of barren plateaus and local minima. Our method is based on coordinate transformations, somehow similar to variational rotations, adding extra directions in parameter space that depend on the cost function itself, and which allow to explore the configuration landscape more efficiently. The validity of our method is benchmarked by boosting a number of quantum machine learning algorithms, getting a very significant improvement in their performance.
title Improving Gradient Methods via Coordinate Transformations: Applications to Quantum Machine Learning
topic Quantum Physics
Machine Learning
url https://arxiv.org/abs/2304.06768