Explainability Requires Interactivity
Not on the map. Why?
Abstract (arXiv)
When explaining the decisions of deep neural networks, simple stories are tempting but dangerous. Especially in computer vision, the most popular explanation approaches give a false sense of comprehension to its users and provide an overly simplistic picture. We introduce an interactive framework to understand the highly complex decision boundaries of modern vision models. It allows the user to exhaustively inspect, probe, and test a network's decisions. Across a range of case studies, we compare the power of our interactive approach to static explanation methods, showing how these can lead a user astray, with potentially severe consequences.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 42 of its 57 references that are in the map. First posted on arXiv 16 Sep 2021.
Builds on
42 of 57 references.
References · 42
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
- 2013Intriguing properties of neural networkscited by 5,155
- 2014Deep Learning Face Attributes in the Wildcited by 3,693
- 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
- 2017Axiomatic Attribution for Deep Networkscited by 3,384
- 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
- 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
- 2013Visualizing and Understanding Convolutional Networkscited by 2,290
- 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
- 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
30 more
- 2018Sanity Checks for Saliency Mapscited by 1,156
- 2017SmoothGrad: removing noise by adding noisecited by 1,109
- 2014Striving for Simplicity: The All Convolutional Netcited by 990
- 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 941
- 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
- 2019MaskGAN: Towards Diverse and Interactive Facial Image Manipulationcited by 628
- 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 612
- 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
- 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 476
- 2019Unmasking Clever Hans Predictors and Assessing What Machines Really Learncited by 365
- 2016Generative Visual Manipulation on the Natural Image Manifoldcited by 348
- 2017The (Un)reliability of saliency methodscited by 333
- 2020StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generationcited by 332
- 2019Counterfactual Visual Explanationscited by 313
- 2016Revisiting Classifier Two-Sample Testscited by 268
- 2017Real Time Image Saliency for Black Box Classifierscited by 246
- 2018Counterfactual Fairness in Text Classification through Robustnesscited by 151
- 2019Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systemscited by 131
- 2019Benchmarking Attribution Methods with Relative Feature Importancecited by 89
- 2020Explainable Artificial Intelligence: a Systematic Reviewcited by 79
- 2019Explanation by Progressive Exaggerationcited by 76
- 2019Image Counterfactual Sensitivity Analysis for Detecting Unintended Biascited by 74
- 2019Perturbation Sensitivity Analysis to Detect Unintended Model Biasescited by 74
- 2020Towards causal benchmarking of bias in face analysis algorithmscited by 64
- 2021Discover the Unknown Biased Attribute of an Image Classifiercited by 46
- 2020Inclusive GAN: Improving Data and Minority Coverage in Generative Modelscited by 44
- 2019Generative Counterfactual Introspection for Explainable Deep Learningcited by 36
- 2021Using StyleGAN for Visual Interpretability of Deep Learning Models on Medical Imagescited by 35
- 2019Two-sample Testing Using Deep Learningcited by 27
- 2018Interactive Classification for Deep Learning Interpretationcited by 2
Built on by
Citing papers per year: 2022: 1.
1 paper.