Saved in:
Bibliographic Details
Main Authors: Shukla, Khemraj, Shin, Yeonjong
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.14168
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914664047181824
author Shukla, Khemraj
Shin, Yeonjong
author_facet Shukla, Khemraj
Shin, Yeonjong
contents We present a randomized forward mode gradient (RFG) as an alternative to backpropagation. RFG is a random estimator for the gradient that is constructed based on the directional derivative along a random vector. The forward mode automatic differentiation (AD) provides an efficient computation of RFG. The probability distribution of the random vector determines the statistical properties of RFG. Through the second moment analysis, we found that the distribution with the smallest kurtosis yields the smallest expected relative squared error. By replacing gradient with RFG, a class of RFG-based optimization algorithms is obtained. By focusing on gradient descent (GD) and Polyak's heavy ball (PHB) methods, we present a convergence analysis of RFG-based optimization algorithms for quadratic functions. Computational experiments are presented to demonstrate the performance of the proposed algorithms and verify the theoretical findings.
format Preprint
id arxiv_https___arxiv_org_abs_2310_14168
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Randomized Forward Mode of Automatic Differentiation For Optimization Algorithms
Shukla, Khemraj
Shin, Yeonjong
Optimization and Control
Artificial Intelligence
Machine Learning
65K05, 65B99, 65Y20
We present a randomized forward mode gradient (RFG) as an alternative to backpropagation. RFG is a random estimator for the gradient that is constructed based on the directional derivative along a random vector. The forward mode automatic differentiation (AD) provides an efficient computation of RFG. The probability distribution of the random vector determines the statistical properties of RFG. Through the second moment analysis, we found that the distribution with the smallest kurtosis yields the smallest expected relative squared error. By replacing gradient with RFG, a class of RFG-based optimization algorithms is obtained. By focusing on gradient descent (GD) and Polyak's heavy ball (PHB) methods, we present a convergence analysis of RFG-based optimization algorithms for quadratic functions. Computational experiments are presented to demonstrate the performance of the proposed algorithms and verify the theoretical findings.
title Randomized Forward Mode of Automatic Differentiation For Optimization Algorithms
topic Optimization and Control
Artificial Intelligence
Machine Learning
65K05, 65B99, 65Y20
url https://arxiv.org/abs/2310.14168