Mercari AI’s research “Want robust explanations? Get smoother predictions first.” accepted to CIKM 2022’s workshop

■Overview

We are pleased to announce that the "Want robust explanations? Get smoother predictions first." by Engineer Deddy Jobson of the Mercaril AI team has been accepted for the AIMLAI workshop of the international conference "CIKM2022".
CIKM (The Conference on Information and Knowledge Management) is one of the most prestigious international conferences in the field of data mining and information retrieval, and is held annually by researchers from around the world.

The accepted papers were presented at the AIMLAI workshop of "CIKM 2022" held at Atlanta, Georgia, USA in October 2022.

■Key points of presentation

• We improve the robustness of model-agnostic ML interpretability methods like LIME and SHAP by smoothening the predictions of the model before passing it through the explainer.
• We find an improvement in the robustness of the explanations as measured by a variant of the Lipshitz score we introduce in our paper.

■Background

• Model-agnostic interpretability methods like LIME and SHAP help explain the predictions made by complex ML algorithms increasing their trustworthiness.
• However, they suffer from a lack of robustness where small perturbations to the input can greatly change the explanations provided by these methods even when the prediction of the output does not change.
• This undermines trust in the explanations and therefore hinders adoption of the ML model.

■Summary of paper

• We propose a simple approach to improve the robustness of the explanations.
• For each data point, we add gaussian noise to the input, pass it through the model, and get the output. We do this multiple times and take the average for each data point.
• The resulting predictions are smoother making the explanations provided by LIME and SHAP more robust and consequently trustworthier.

■About MDS Team

The Marketing Data Science Team makes use of statistics, machine learning, and mathematical optimization to improve the efficiency of marketing campaigns. We actively learn and adopt the state of the art methods used in the literature to solve technical problems and publish papers along the way.