mlsanityData to 7 Oct 2026

NoiseGrad: Enhancing Explanations by Introducing Stochasticity to Model Weights

arXiv 2106.10185 ↗First posted 18 Jun 2021cs.LG cs.AIcited by 28

Not on the map. Why?

Abstract (arXiv)

Many efforts have been made for revealing the decision-making process of black-box learning machines such as deep neural networks, resulting in useful local and global explanation methods. For local explanation, stochasticity is known to help: a simple method, called SmoothGrad, has improved the visual quality of gradient-based attribution by adding noise to the input space and averaging the explanations of the noisy inputs. In this paper, we extend this idea and propose NoiseGrad that enhances both local and global explanation methods. Specifically, NoiseGrad introduces stochasticity in the weight parameter space, such that the decision boundary is perturbed. NoiseGrad is expected to enhance the local explanation, similarly to SmoothGrad, due to the dual relationship between the input perturbation and the decision boundary perturbation. We evaluate NoiseGrad and its fusion with SmoothGrad -- FusionGrad -- qualitatively and quantitatively with several evaluation criteria, and show that our novel approach significantly outperforms the baseline methods. Both NoiseGrad and FusionGrad are method-agnostic and as handy as SmoothGrad using a simple heuristic for the choice of the hyperparameter setting without the need of finetuning.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). Builds on 41 of its 86 references that are in the map. First posted on arXiv 18 Jun 2021.

Builds on

41 of 86 references.

References · 41

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  6. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  7. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  8. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  9. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  10. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  11. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  12. 2018Sanity Checks for Saliency Mapscited by 1,149
29 more
  1. 2017SmoothGrad: removing noise by adding noisecited by 1,097
  2. 2018A Survey Of Methods For Explaining Black Box Modelscited by 970
  3. 2015Optimizing Neural Networks with Kronecker-factored Approximate Curvaturecited by 776
  4. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  5. 2015Variational Dropout and the Local Reparameterization Trickcited by 489
  6. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  7. 2015Evaluating the visualization of what a Deep Neural Network has learnedcited by 416
  8. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 342
  9. 2017The (Un)reliability of saliency methodscited by 331
  10. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  11. 2019On the (In)fidelity and Sensitivity for Explanationscited by 309
  12. 2017Variational Dropout Sparsifies Deep Neural Networkscited by 265
  13. 2019Explanations can be manipulated and geometry is to blamecited by 215
  14. 2020Evaluating and Aggregating Feature-based Model Explanationscited by 185
  15. 2016Synthesizing the preferred inputs for neurons in neural networks via deep generator networkscited by 181
  16. 2019Practical Deep Learning with Bayesian Principlescited by 158
  17. 2019Benchmarking Attribution Methods with Relative Feature Importancecited by 89
  18. 2020On quantitative aspects of model interpretabilitycited by 71
  19. 2019Towards Best Practice in Explaining Neural Network Decisions with LRPcited by 70
  20. 2020Scaling Hamiltonian Monte Carlo Inference for Bayesian Neural Networks with Symmetric Splittingcited by 54
  21. 2016"What is Relevant in a Text Document?": An Interpretable Machine Learning Approachcited by 53
  22. 2019Robust Attribution Regularizationcited by 46
  23. 2020Estimating Model Uncertainty of Neural Networks in Sparse Information Formcited by 31
  24. 2014Analyzing noise in autoencoders and deep networkscited by 25
  25. 2015Analyzing Classifiers: Fisher Vectors and Deep Neural Networkscited by 24
  26. 2020Black Box Explanation by Learning Image Exemplars in the Latent Feature Spacecited by 23
  27. 2020A simple defense against adversarial attacks on heatmap explanationscited by 23
  28. 2020How Much Can I Trust You? -- Quantifying Uncertainties in Explaining Neural Networkscited by 18
  29. 2016Feature Importance Measure for Non-linear Learning Algorithmscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 7; 2023: 3; 2024: 10; 2025: 4; 2026: 3.

28 papers.

Sort:

  1. 2022Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyondcited by 171
  2. 2023The Meta-Evaluation Problem in Explainable AI: Identifying Reliable Estimators with MetaQuantuscited by 37
  3. 2023Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Sciencecited by 25
  4. 2022What Makes a Good Explanation?: A Harmonized View of Properties of Explanationscited by 21
  5. 2021Explaining Bayesian Neural Networkscited by 18
  6. 2022DORA: Exploring Outlier Representations in Deep Neural Networkscited by 14
  7. 2022SInGE: Sparsity via Integrated Gradients Estimation of Neuron Relevancecited by 9
  8. 2024Sanity Checks Revisited: An Exploration to Repair the Model Parameter Randomisation Testcited by 9
  9. 2024A Fresh Look at Sanity Checks for Saliency Mapscited by 8
  10. 2022SoK: Modeling Explainability in Security Analytics for Interpretability, Trustworthiness, and Usabilitycited by 7
  11. 2024Provably Better Explanations with Optimized Aggregation of Feature Attributionscited by 6
  12. 2022Visualizing the Diversity of Representations Learned by Bayesian Neural Networkscited by 4
16 more
  1. 2024From Flexibility to Manipulation: The Slippery Slope of XAI Evaluationcited by 4
  2. 2024GARNN: An Interpretable Graph Attentive Recurrent Neural Network for Predicting Blood Glucose Levels via Multivariate Time Seriescited by 2
  3. 2024PASA: Attack Agnostic Unsupervised Adversarial Detection using Prediction & Attribution Sensitivity Analysiscited by 2
  4. 2024AdaptGrad: Adaptive Sampling to Reduce Noisecited by 2
  5. 2025On the Complexity-Faithfulness Trade-off of Gradient-Based Explanationscited by 2
  6. 2025On Spectral Properties of Gradient-based Explanation Methodscited by 2
  7. 2022Multi-Label Classification of Thoracic Diseases using Dense Convolutional Network on Chest Radiographscited by 1
  8. 2023The Paradox of Noise: An Empirical Study of Noise-Infusion Mechanisms to Improve Generalization, Stability, and Privacy in Federated Learningcited by 1
  9. 2024CNN-based explanation ensembling for dataset, representation and explanations evaluationcited by 1
  10. 2024Beyond the Black Box: Do More Complex Deep Learning Models Provide Superior XAI Explanations?cited by 1
  11. 2024Axiomatization of Gradient Smoothing in Neural Networkscited by 0
  12. 2025Pulling Back the Curtain on Deep Networkscited by 0
  13. 2025Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruningcited by 0
  14. 2026Training for Trustworthy Saliency Maps: Adversarial Training Meets Feature-Map Smoothingcited by 0
  15. 2026Explanation of Dynamic Physical Field Predictions using WassersteinGrad: Application to Autoregressive Weather Forecastingcited by 0
  16. 2026Scaling Vision Models Does Not Consistently Improve Localisation-Based Explanation Qualitycited by 0