# Better sampling in explanation methods can prevent dieselgate-like deception

arXiv: https://arxiv.org/abs/2101.11702 · HTML: https://mlsanity.com/paper/2101.11702/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 30 references that are in the map.
- First posted on arXiv 26 Jan 2021.

## Builds on (most cited first, 13 of 13)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [The Mythos of Model Interpretability](https://mlsanity.com/paper/1606.03490/) (2016)
- [Interpretation of Neural Networks is Fragile](https://mlsanity.com/paper/1710.10547/) (2017)
- [Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods](https://mlsanity.com/paper/1911.02508/) (2019)
- [Tutorial on Variational Autoencoders](https://mlsanity.com/paper/1606.05908/) (2016)
- [On the Robustness of Interpretability Methods](https://mlsanity.com/paper/1806.08049/) (2018)
- [Explanations can be manipulated and geometry is to blame](https://mlsanity.com/paper/1906.07983/) (2019)
- [Fooling Neural Network Interpretations via Adversarial Model Manipulation](https://mlsanity.com/paper/1902.02041/) (2019)
- [Making Fair ML Software using Trustworthy Explanation](https://mlsanity.com/paper/2007.02893/) (2020)
- [Improving LIME Robustness with Smarter Locality Sampling](https://mlsanity.com/paper/2006.12302/) (2020)
- [Generating Data using Monte Carlo Dropout](https://mlsanity.com/paper/1909.05755/) (2019)

## Built on by (most cited first, 5 of 5)

- [Don't Lie to Me: Avoiding Malicious Explanations with STEALTH](https://mlsanity.com/paper/2301.10407/) (2023)
- [TrustyAI Explainability Toolkit](https://mlsanity.com/paper/2104.12717/) (2021)
- [On the Bias-Variance Characteristics of LIME and SHAP in High Sparsity Movie Recommendation Explanation Tasks](https://mlsanity.com/paper/2206.04784/) (2022)
- [EMaP: Explainable AI with Manifold-based Perturbations](https://mlsanity.com/paper/2209.08453/) (2022)
- [MindfulLIME: A Stable Solution for Explanations of Machine Learning Models with Enhanced Localization Precision -- A Medical Image Case Study](https://mlsanity.com/paper/2503.20758/) (2025)

## Abstract (arXiv)

Machine learning models are used in many sensitive areas where besides predictive accuracy their comprehensibility is also important. Interpretability of prediction models is necessary to determine their biases and causes of errors, and is a necessary prerequisite for users' confidence. For complex state-of-the-art black-box models post-hoc model-independent explanation techniques are an established solution. Popular and effective techniques, such as IME, LIME, and SHAP, use perturbation of instance features to explain individual predictions. Recently, Slack et al. (2020) put their robustness into question by showing that their outcomes can be manipulated due to poor perturbation sampling employed. This weakness would allow dieselgate type cheating of owners of sensitive models who could deceive inspection and hide potentially unethical or illegal biases existing in their predictive models. This could undermine public trust in machine learning models and give rise to legal restrictions on their use. We show that better sampling in these explanation methods prevents malicious manipulations. The proposed sampling uses data generators that learn the training set distribution and generate new perturbation instances much more similar to the training set. We show that the improved sampling increases the robustness of the LIME and SHAP, while previously untested method IME is already the most robust of all.