# Decision-Directed Data Decomposition

arXiv: https://arxiv.org/abs/1909.08159 · HTML: https://mlsanity.com/paper/1909.08159/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Sep 2019.

## Built on by (most cited first, 2 of 2)

- [Marked Attribute Bias in Natural Language Inference](https://mlsanity.com/paper/2109.14039/) (2021)
- [Second Order WinoBias (SoWinoBias) Test Set for Latent Gender Bias Detection in Coreference Resolution](https://mlsanity.com/paper/2109.14047/) (2021)

## Abstract (arXiv)

We present an algorithm, Decision-Directed Data Decomposition (D4), which decomposes a dataset into two components. The first contains most of the useful information for a specified supervised learning task. The second orthogonal component contains little information about the task but retains associations and information that were not targeted. The algorithm is simple and scalable. We illustrate its application in image and text processing domains. Our results show that 1) post-hoc application of D4 to an image representation space can remove information about specified concepts without impacting other concepts, 2) D4 is able to improve predictive generalization in certain settings, and 3) applying D4 to word embedding representations produces state-of-the-art results in debiasing.