AIM is a new saliency-guided adversarial feature replacement method to evaluate faithfulness of saliency maps and reliability of masking operators on image, audio, and EEG tasks.
Towards better understanding of gradient-based attribution methods for Deep Neural Networks
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
Understanding the flow of information in Deep Neural Networks (DNNs) is a challenging problem that has gain increasing attention over the last few years. While several methods have been proposed to explain network predictions, there have been only a few attempts to compare them from a theoretical perspective. What is more, no exhaustive empirical comparison has been performed in the past. In this work, we analyze four gradient-based attribution methods and formally prove conditions of equivalence and approximation between them. By reformulating two of these methods, we construct a unified framework which enables a direct comparison, as well as an easier implementation. Finally, we propose a novel evaluation metric, called Sensitivity-n and test the gradient-based attribution methods alongside with a simple perturbation-based attribution method on several datasets in the domains of image and text classification, using various network architectures.
verdicts
UNVERDICTED 3representative citing papers
CAAF clusters candidate sensor locations before applying feature attribution to reduce redundancy and improve optimal sensor placement for predictions in dynamical systems.
WSVD delivers over 1.8x faster VLM decoding via weighted low-rank approximation at fine granularity plus quantization, without accuracy loss.
citing papers explorer
-
AIM: Adversarial Information Masking for Faithfulness Evaluation of Saliency Maps
AIM is a new saliency-guided adversarial feature replacement method to evaluate faithfulness of saliency maps and reliability of masking operators on image, audio, and EEG tasks.
-
Data-driven Sensor Placement for Predictive Applications: A Correlation-Assisted Attribution Framework (CAAF)
CAAF clusters candidate sensor locations before applying feature attribution to reduce redundancy and improve optimal sensor placement for predictions in dynamical systems.
-
WSVD: Weighted Low-Rank Approximation for Fast and Efficient Execution of Low-Precision Vision-Language Models
WSVD delivers over 1.8x faster VLM decoding via weighted low-rank approximation at fine granularity plus quantization, without accuracy loss.