Accelerating Matrix Diagonalization through Decision Transformers with Epsilon-Greedy Optimization

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bhatta, Kshitij, Zollicoffer, Geigh, Bhattarai, Manish, Romero, Phil, Negre, Christian F. A., Niklasson, Anders M. N., Adedoyin, Adetokunbo
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909230093565952
author Bhatta, Kshitij
Zollicoffer, Geigh
Bhattarai, Manish
Romero, Phil
Negre, Christian F. A.
Niklasson, Anders M. N.
Adedoyin, Adetokunbo
author_facet Bhatta, Kshitij
Zollicoffer, Geigh
Bhattarai, Manish
Romero, Phil
Negre, Christian F. A.
Niklasson, Anders M. N.
Adedoyin, Adetokunbo
contents This paper introduces a novel framework for matrix diagonalization, recasting it as a sequential decision-making problem and applying the power of Decision Transformers (DTs). Our approach determines optimal pivot selection during diagonalization with the Jacobi algorithm, leading to significant speedups compared to the traditional max-element Jacobi method. To bolster robustness, we integrate an epsilon-greedy strategy, enabling success in scenarios where deterministic approaches fail. This work demonstrates the effectiveness of DTs in complex computational tasks and highlights the potential of reimagining mathematical operations through a machine learning lens. Furthermore, we establish the generalizability of our method by using transfer learning to diagonalize matrices of smaller sizes than those trained.
format Preprint
id arxiv_https___arxiv_org_abs_2406_16191
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Accelerating Matrix Diagonalization through Decision Transformers with Epsilon-Greedy Optimization
Bhatta, Kshitij
Zollicoffer, Geigh
Bhattarai, Manish
Romero, Phil
Negre, Christian F. A.
Niklasson, Anders M. N.
Adedoyin, Adetokunbo
Machine Learning
Artificial Intelligence
Numerical Analysis
This paper introduces a novel framework for matrix diagonalization, recasting it as a sequential decision-making problem and applying the power of Decision Transformers (DTs). Our approach determines optimal pivot selection during diagonalization with the Jacobi algorithm, leading to significant speedups compared to the traditional max-element Jacobi method. To bolster robustness, we integrate an epsilon-greedy strategy, enabling success in scenarios where deterministic approaches fail. This work demonstrates the effectiveness of DTs in complex computational tasks and highlights the potential of reimagining mathematical operations through a machine learning lens. Furthermore, we establish the generalizability of our method by using transfer learning to diagonalize matrices of smaller sizes than those trained.
title Accelerating Matrix Diagonalization through Decision Transformers with Epsilon-Greedy Optimization
topic Machine Learning
Artificial Intelligence
Numerical Analysis
url https://arxiv.org/abs/2406.16191