# Randomized Ablation Feature Importance

arXiv: https://arxiv.org/abs/1910.00174 · HTML: https://mlsanity.com/paper/1910.00174/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Oct 2019.

## Built on by (most cited first, 3 of 3)

- [Towards Explainable Artificial Intelligence (XAI): A Data Mining Perspective](https://mlsanity.com/paper/2401.04374/) (2024)
- [Neural Insights for Digital Marketing Content Design](https://mlsanity.com/paper/2302.01416/) (2023)
- [Transforming gradient-based techniques into interpretable methods](https://mlsanity.com/paper/2401.14434/) (2024)

## Abstract (arXiv)

Given a model f that predicts a target y from a vector of input features pmbx = x1, x2, ldots, xM, we seek to measure the importance of each feature with respect to the model's ability to make a good prediction. To this end, we consider how (on average) some measure of goodness or badness of prediction (which we term "loss" ℓ), changes when we hide or ablate each feature from the model. To ablate a feature, we replace its value with another possible value randomly. By averaging over many points and many possible replacements, we measure the importance of a feature on the model's ability to make good predictions. Furthermore, we present statistical measures of uncertainty that quantify how confident we are that the feature importance we measure from our finite dataset and finite number of ablations is close to the theoretical true importance value.