Learning Propagation Rules for Attribution Map Generation

10/14/2020
by   Yiding Yang, et al.
0

Prior gradient-based attribution-map methods rely on handcrafted propagation rules for the non-linear/activation layers during the backward pass, so as to produce gradients of the input and then the attribution map. Despite the promising results achieved, such methods are sensitive to the non-informative high-frequency components and lack adaptability for various models and samples. In this paper, we propose a dedicated method to generate attribution maps that allow us to learn the propagation rules automatically, overcoming the flaws of the handcrafted ones. Specifically, we introduce a learnable plugin module, which enables adaptive propagation rules for each pixel, to the non-linear layers during the backward pass for mask generating. The masked input image is then fed into the model again to obtain new output that can be used as a guidance when combined with the original one. The introduced learnable module can be trained under any auto-grad framework with higher-order differential support. As demonstrated on five datasets and six network architectures, the proposed method yields state-of-the-art results and gives cleaner and more visually plausible attribution maps.

READ FULL TEXT

page 1

page 2

page 3

page 4

01/17/2021

Generating Attribution Maps with Disentangled Masked Backpropagation

Attribution map visualization has arisen as one of the most effective te...
05/30/2022

CHALLENGER: Training with Attribution Maps

We show that utilizing attribution maps for training neural networks can...
12/07/2020

Interpreting Deep Neural Networks with Relative Sectional Propagation by Analyzing Comparative Gradients and Hostile Activations

The clear transparency of Deep Neural Networks (DNNs) is hampered by com...
02/15/2021

Attribution Mask: Filtering Out Irrelevant Features By Recursively Focusing Attention on Inputs of DNNs

Attribution methods calculate attributions that visually explain the pre...
03/22/2022

Exploring Linear Feature Disentanglement For Neural Networks

Non-linear activation functions, e.g., Sigmoid, ReLU, and Tanh, have ach...
02/18/2020

Camera Model Anonymisation with Augmented cGANs

The model of camera that was used to capture a particular photographic i...
04/13/2022

OccAM's Laser: Occlusion-based Attribution Maps for 3D Object Detectors on LiDAR Data

While 3D object detection in LiDAR point clouds is well-established in a...