Aggregating explanation methods for stable and robust explainability

2019/03/01 by Laura Rieger, Lars Kai Hansen, Rieger, Laura +1 · 3 citations
Computer Science · Mathematics · Medicine · #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #Artificial Intelligence in Healthcare and Education #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #cs.AI #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1903.00519

openalex publication_date 2019/03/01 · arxiv created 2020/03/20 · arxiv updated 2020/03/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Despite a growing literature on explaining neural networks, no consensus has been reached on how to explain a neural network decision or how to evaluate an explanation. Our contributions in this paper are twofold. First, we investigate schemes to combine explanation methods and reduce model uncertainty to obtain a single aggregated explanation. We provide evidence that the aggregation is better at identifying important features, than on individual methods. Adversarial attacks on explanations is a recent active research topic. As our second contribution, we present evidence that aggregate explanations are much more robust to attacks than individual explanation methods.

Citations

• ADADELTA: An Adaptive Learning Rate Method
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Neural Networks and the Bias/Variance Dilemma
• Axiomatic Attribution for Deep Networks
• SmoothGrad: removing noise by adding noise
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• A Human-Grounded Evaluation Benchmark for Local Explanations of Machine Learning
• Sanity Checks for Saliency Maps
• Interpretable Deep Learning under Fire
• Fairwashing: the risk of rationalization
• Explanations can be manipulated and geometry is to blame
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
• On a Method to Measure Supervised Multiclass Model’s Interpretability: Application to Degradation Diagnosis (Short Paper)

Cited by

• Explainable AI for online disinformation detection: Insights from a design science research project
• Aggregating Local Saliency Maps for Semi-Global Explainable Image Classification
• Multi-criteria Rank-based Aggregation for Explainable AI

Related

• Do Explanations Explain? Model Knows Best
• When Can You Trust Your Explanations? A Robustness Analysis on Feature Importances
• Towards a Unified Evaluation of Explanation Methods without Ground Truth
• Explanations can be manipulated and geometry is to blame
• Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations