# Unifying Perplexing Behaviors in Modified BP Attributions through Alignment Perspective

arXiv: https://arxiv.org/abs/2503.11160 · HTML: https://mlsanity.com/paper/2503.11160/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 34 of its 41 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Builds on (most cited first, 30 of 34)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Sanity Checks for Saliency Maps](https://mlsanity.com/paper/1810.03292/) (2018)
- [BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain](https://mlsanity.com/paper/1708.06733/) (2017)
- [Striving for Simplicity: The All Convolutional Net](https://mlsanity.com/paper/1412.6806/) (2014)
- [Robustness May Be at Odds with Accuracy](https://mlsanity.com/paper/1805.12152/) (2018)
- [Evaluating the visualization of what a Deep Neural Network has learned](https://mlsanity.com/paper/1509.06321/) (2015)
- [A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI](https://mlsanity.com/paper/1907.07374/) (2019)
- [Unmasking Clever Hans Predictors and Assessing What Machines Really Learn](https://mlsanity.com/paper/1902.10178/) (2019)
- [Top-down Neural Attention by Excitation Backprop](https://mlsanity.com/paper/1608.00507/) (2016)
- [Input-Aware Dynamic Backdoor Attack](https://mlsanity.com/paper/2010.08138/) (2020)
- [A Survey on Neural Network Interpretability](https://mlsanity.com/paper/2012.14261/) (2020)
- [Explaining NonLinear Classification Decisions with Deep Taylor Decomposition](https://mlsanity.com/paper/1512.02479/) (2015)
- [Evaluating and Aggregating Feature-based Model Explanations](https://mlsanity.com/paper/2005.00631/) (2020)
- [Dynamic Backdoor Attacks Against Machine Learning Models](https://mlsanity.com/paper/2003.03675/) (2020)
- [Interpretations are useful: penalizing explanations to align neural networks with prior knowledge](https://mlsanity.com/paper/1909.13584/) (2019)
- [When Explanations Lie: Why Many Modified BP Attributions Fail](https://mlsanity.com/paper/1912.09818/) (2019)
- [On the Connection Between Adversarial Robustness and Saliency Map Interpretability](https://mlsanity.com/paper/1905.04172/) (2019)
- [Attribution in Scale and Space](https://mlsanity.com/paper/2004.03383/) (2020)
- [A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations](https://mlsanity.com/paper/1805.07039/) (2018)
- [There and Back Again: Revisiting Backpropagation Saliency Methods](https://mlsanity.com/paper/2004.02866/) (2020)
- [Local Explanation Methods for Deep Neural Networks Lack Sensitivity to Parameter Values](https://mlsanity.com/paper/1810.03307/) (2018)
- [Why are Saliency Maps Noisy? Cause of and Solution to Noisy Saliency Maps](https://mlsanity.com/paper/1902.04893/) (2019)
- [What Do You See? Evaluation of Explainable Artificial Intelligence (XAI) Interpretability through Neural Backdoors](https://mlsanity.com/paper/2009.10639/) (2020)
- [Bridging Adversarial Robustness and Gradient Interpretability](https://mlsanity.com/paper/1903.11626/) (2019)
- [Shortcomings of Top-Down Randomization-Based Sanity Checks for Evaluations of Deep Neural Network Explanations](https://mlsanity.com/paper/2211.12486/) (2022)

## Abstract (arXiv)

Attributions aim to identify input pixels that are relevant to the decision-making process. A popular approach involves using modified backpropagation (BP) rules to reverse decisions, which improves interpretability compared to the original gradients. However, these methods lack a solid theoretical foundation and exhibit perplexing behaviors, such as reduced sensitivity to parameter randomization, raising concerns about their reliability and highlighting the need for theoretical justification. In this work, we present a unified theoretical framework for methods like GBP, RectGrad, LRP, and DTD, demonstrating that they achieve input alignment by combining the weights of activated neurons. This alignment improves the visualization quality and reduces sensitivity to weight randomization. Our contributions include: (1) Providing a unified explanation for multiple behaviors, rather than focusing on just one. (2) Accurately predicting novel behaviors. (3) Offering insights into decision-making processes, including layer-wise information changes and the relationship between attributions and model decisions.