# Attribution in Scale and Space

arXiv: https://arxiv.org/abs/2004.03383 · HTML: https://mlsanity.com/paper/2004.03383/

## Key facts

- Cited by 61 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Apr 2020.

## Built on by (most cited first, 30 of 61)

- [Explaining in Style: Training a GAN to explain a classifier in StyleSpace](https://mlsanity.com/paper/2104.13369/) (2021)
- [Guided Integrated Gradients: An Adaptive Path Method for Removing Noise](https://mlsanity.com/paper/2106.09788/) (2021)
- [A Consistent and Efficient Evaluation Strategy for Attribution Methods](https://mlsanity.com/paper/2202.00449/) (2022)
- [WaveFake: A Data Set to Facilitate Audio Deepfake Detection](https://mlsanity.com/paper/2111.02813/) (2021)
- [A Rigorous Study of Integrated Gradients Method and Extensions to Internal Neuron Attributions](https://mlsanity.com/paper/2202.11912/) (2022)
- [Learning to Estimate Shapley Values with Vision Transformers](https://mlsanity.com/paper/2206.05282/) (2022)
- [Group-CAM: Group Score-Weighted Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/2103.13859/) (2021)
- [Gradient based Feature Attribution in Explainable AI: A Technical Review](https://mlsanity.com/paper/2403.10415/) (2024)
- [Explainable Person Re-Identification with Attribute-guided Metric Distillation](https://mlsanity.com/paper/2103.01451/) (2021)
- [Manifold Integrated Gradients: Riemannian Geometry for Feature Attribution](https://mlsanity.com/paper/2405.09800/) (2024)
- [Visual Explanations via Iterated Integrated Attributions](https://mlsanity.com/paper/2310.18585/) (2023)
- [IDGI: A Framework to Eliminate Explanation Noise from Integrated Gradients](https://mlsanity.com/paper/2303.14242/) (2023)
- [On the Relationship Between Explanation and Prediction: A Causal View](https://mlsanity.com/paper/2212.06925/) (2022)
- [IG2: Integrated Gradient on Iterative Gradient Path for Feature Attribution](https://mlsanity.com/paper/2406.10852/) (2024)
- [Probing Deep into Temporal Profile Makes the Infrared Small Target Detector Much Better](https://mlsanity.com/paper/2506.12766/) (2025)
- [Deep Integrated Explanations](https://mlsanity.com/paper/2310.15368/) (2023)
- [MAEA: Multimodal Attribution for Embodied AI](https://mlsanity.com/paper/2307.13850/) (2023)
- [Attribution of Predictive Uncertainties in Classification Models](https://mlsanity.com/paper/2107.08756/) (2021)
- [Gradient-based Uncertainty Attribution for Explainable Bayesian Deep Learning](https://mlsanity.com/paper/2304.04824/) (2023)
- [Attention Mechanisms Don't Learn Additive Models: Rethinking Feature Importance for Transformers](https://mlsanity.com/paper/2405.13536/) (2024)
- [Do Different Deep Metric Learning Losses Lead to Similar Learned Features?](https://mlsanity.com/paper/2205.02698/) (2022)
- [Trade-off Between Efficiency and Consistency for Removal-based Explanations](https://mlsanity.com/paper/2210.17426/) (2022)
- [Four Axiomatic Characterizations of the Integrated Gradients Attribution Method](https://mlsanity.com/paper/2306.13753/) (2023)
- [DiffExplainer: Unveiling Black Box Models Via Counterfactual Generation](https://mlsanity.com/paper/2406.15182/) (2024)
- [Path Choice Matters for Clear Attribution in Path Methods](https://mlsanity.com/paper/2401.10442/) (2024)
- [Reliable Evaluation of Attribution Maps in CNNs: A Perturbation-Based Approach](https://mlsanity.com/paper/2411.14946/) (2024)
- [Using the Path of Least Resistance to Explain Deep Networks](https://mlsanity.com/paper/2502.12108/) (2025)
- [Spatio-Temporal Perturbations for Video Attribution](https://mlsanity.com/paper/2109.00222/) (2021)
- [Comparing Baseline Shapley and Integrated Gradients for Local Explanation: Some Additional Insights](https://mlsanity.com/paper/2208.06096/) (2022)
- [Quantitative Metrics for Evaluating Explanations of Video DeepFake Detectors](https://mlsanity.com/paper/2210.03683/) (2022)

## Abstract (arXiv)

We study the attribution problem [28] for deep networks applied to perception tasks. For vision tasks, attribution techniques attribute the prediction of a network to the pixels of the input image. We propose a new technique called Blur Integrated Gradients. This technique has several advantages over other methods. First, it can tell at what scale a network recognizes an object. It produces scores in the scale/frequency dimension, that we find captures interesting phenomena. Second, it satisfies the scale-space axioms [14], which imply that it employs perturbations that are free of artifact. We therefore produce explanations that are cleaner and consistent with the operation of deep networks. Third, it eliminates the need for a 'baseline' parameter for Integrated Gradients [31] for perception tasks. This is desirable because the choice of baseline has a significant effect on the explanations. We compare the proposed technique against previous techniques and demonstrate application on three tasks: ImageNet object recognition, Diabetic Retinopathy prediction, and AudioSet audio event identification.