Applying Adaptive Gradient Descent to solve matrix factorization

TL;DR

This paper introduces an adaptive gradient descent method for matrix factorization that adjusts step lengths over epochs, outperforming traditional fixed gradient descent while maintaining convergence guarantees.

Contribution

The paper proposes a novel adaptive gradient descent approach for matrix factorization that improves performance over fixed gradient methods.

Findings

Adaptive gradient descent outperforms fixed gradient descent in tests.

The method maintains convergence guarantees.

Performance improvements are consistent across experiments.

Abstract

Based on the method of FGD, we apply the method of adaptive gradient descent which uses different step length at different epoch. Adaptive gradient descent performs much better than FGD in the tests and keeps the guarantee of convergence speed at the same time.