paper-with-me

Papers

Randomized Forward Mode of Automatic Differentiation For Optimization Algorithms

2023-10-22 · Khemraj Shukla, Yeonjong Shin

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.

📄 PDF Abstract BibTeX arXiv:2310.14168

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Collapsing Taylor Mode Automatic Differentiation

2025-05-19 · Felix Dangel, Tim Siebert, Marius Zeinhofer, Andrea Walther

Computing partial differential equation (PDE) operators via nested backpropagation is expensive, yet popular, and severely restricts their utility for scientific machine learning. Recent advances, like the forward Laplac…

Differentiation Strategies for Acoustic Inverse Problems: Admittance Estimation and Shape Optimization

2025-11-14 · Nikolas Borrel-Jensen, Josiah Bjorgaard arxiv

We demonstrate a practical differentiable programming approach for acoustic inverse problems through two applications: admittance estimation and shape optimization for resonance damping. First, we show that JAX-FEM's aut…

Randomized Automatic Differentiation

2020-07-20 · ICLR 2021 1 · Deniz Oktay, Nick McGreivy, Joshua Aduol, Alex Beatson 외

The successes of deep learning, variational inference, and many other fields have been aided by specialized implementations of reverse-mode automatic differentiation (AD) to compute gradients of mega-dimensional objectiv…

Stochastic OptimizationVariational Inference

Randomized Subspace Nesterov Accelerated Gradient

2026-05-01 · Gaku Omiya, Pierre-Louis Poirion, Akiko Takeda arxiv

Randomized-subspace methods reduce the cost of first-order optimization by using only low-dimensional projected-gradient information, a feature that is attractive in forward-mode automatic differentiation and communicati…

Source-to-Source Automatic Differentiation of OpenMP Parallel Loops

2021-11-02 · Jan Hückelheim, Laurent Hascoët

This paper presents our work toward correct and efficient automatic differentiation of OpenMP parallel worksharing loops in forward and reverse mode. Automatic differentiation is a method to obtain gradients of numerical…

Uncertainty Quantification