Extending the Forward Forward Algorithm

TL;DR

This paper extends the Forward Forward algorithm beyond vision tasks to sentiment analysis, introduces a pyramidal hyperparameter optimization, and provides insights into trained network parameters, demonstrating improved performance and understanding.

Contribution

The paper is the first to apply the Forward Forward algorithm to NLP sentiment analysis and proposes a pyramidal threshold optimization strategy.

Findings

Achieved baseline performance on IMDb sentiment analysis.

Pyramidal threshold optimization reduces test error by up to 8%.

Forward Forward networks exhibit significantly larger weights than traditional networks.

Abstract

The Forward Forward algorithm, proposed by Geoffrey Hinton in November 2022, is a novel method for training neural networks as an alternative to backpropagation. In this project, we replicate Hinton's experiments on the MNIST dataset, and subsequently extend the scope of the method with two significant contributions. First, we establish a baseline performance for the Forward Forward network on the IMDb movie reviews dataset. As far as we know, our results on this sentiment analysis task marks the first instance of the algorithm's extension beyond computer vision. Second, we introduce a novel pyramidal optimization strategy for the loss threshold - a hyperparameter specific to the Forward Forward method. Our pyramidal approach shows that a good thresholding strategy causes a difference of up to 8% in test error. Lastly, we perform visualizations of the trained parameters and derived several significant insights, such as a notably larger (10-20x) mean and variance in the weights acquired by the Forward Forward network. Repository: https://github.com/Ads-cmu/ForwardForward