Exploring Interpretability of Independent Components of Word Embeddings with Automated Word Intruder Test

TL;DR

This paper investigates the interpretability of word embedding components using Independent Component Analysis (ICA) and introduces an automated word intruder test to evaluate semantic features efficiently.

Contribution

It demonstrates that ICA can uncover meaningful semantic features in word embeddings and proposes an automated intruder test for rapid interpretability assessment.

Findings

ICA reveals semantic features in word embeddings

Most independent components correspond to meaningful features

Automated intruder test effectively quantifies interpretability

Abstract

Independent Component Analysis (ICA) is an algorithm originally developed for finding separate sources in a mixed signal, such as a recording of multiple people in the same room speaking at the same time. Unlike Principal Component Analysis (PCA), ICA permits the representation of a word as an unstructured set of features, without any particular feature being deemed more significant than the others. In this paper, we used ICA to analyze word embeddings. We have found that ICA can be used to find semantic features of the words, and these features can easily be combined to search for words that satisfy the combination. We show that most of the independent components represent such features. To quantify the interpretability of the components, we use the word intruder test, performed both by humans and by large language models. We propose to use the automated version of the word intruder test as a fast and inexpensive way of quantifying vector interpretability without the need for human effort.