↓Skip to main content

Nesterov's Acceleration Gradient Descent Method

1 min·
Table of Contents

This is my STAT3007 tutorial paper that I wrote in Semester 1, 2024.

In this tutorial paper, we will introduce Nesterov’s Accelerated Gradient Descent (AGD) method, which uses acceleration- and momentum-based concepts to overcome the limitations found in GD. Before doing so, we will cover the GD and classical momentum methods, as these are prerequisite algorithms for AGD, and also serve as algorithms we can compare against AGD.

  • Report link: pdf