Blind Descent: A Prequel to Gradient Descent

TL;DR

This paper introduces Blind Descent, an alternative neural network training method that does not rely on gradients, potentially overcoming issues like vanishing or exploding gradients, and demonstrates its feasibility through experiments.

Contribution

The paper presents Blind Descent as a fundamental learning process and shows how gradient descent is a special case within this framework.

Findings

Blind Descent avoids gradient-related problems.

Gradient descent is a specific case of Blind Descent.

Proof of concept demonstrated on neural networks.

Abstract

We describe an alternative learning method for neural networks, which we call Blind Descent. By design, Blind Descent does not face problems like exploding or vanishing gradients. In Blind Descent, gradients are not used to guide the learning process. In this paper, we present Blind Descent as a more fundamental learning process compared to gradient descent. We also show that gradient descent can be seen as a specific case of the Blind Descent algorithm. We also train two neural network architectures, a multilayer perceptron and a convolutional neural network, using the most general Blind Descent algorithm to demonstrate a proof of concept.