DeepAI AI Chat
Log In Sign Up

Learning Robust Convolutional Neural Networks with Relevant Feature Focusing via Explanations

02/09/2022
by   Kazuki Adachi, et al.
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
0

Existing image recognition techniques based on convolutional neural networks (CNNs) basically assume that the training and test datasets are sampled from i.i.d distributions. However, this assumption is easily broken in the real world because of the distribution shift that occurs when the co-occurrence relations between objects and backgrounds in input images change. Under this type of distribution shift, CNNs learn to focus on features that are not task-relevant, such as backgrounds from the training data, and degrade their accuracy on the test data. To tackle this problem, we propose relevant feature focusing (ReFF). ReFF detects task-relevant features and regularizes CNNs via explanation outputs (e.g., Grad-CAM). Since ReFF is composed of post-hoc explanation modules, it can be easily applied to off-the-shelf CNNs. Furthermore, ReFF requires no additional inference cost at test time because it is only used for regularization while training. We demonstrate that CNNs trained with ReFF focus on features relevant to the target task and that ReFF improves the test-time accuracy.

READ FULL TEXT

page 4

page 9

page 10

page 11

page 12

02/10/2022

A Field of Experts Prior for Adapting Neural Networks at Test Time

Performance of convolutional neural networks (CNNs) in image analysis ta...
10/18/2018

Automatic Brain Tumor Segmentation using Convolutional Neural Networks with Test-Time Augmentation

Automatic brain tumor segmentation plays an important role for diagnosis...
03/06/2023

Testing the Channels of Convolutional Neural Networks

Neural networks have complex structures, and thus it is hard to understa...
04/06/2021

Shapley Explanation Networks

Shapley values have become one of the most popular feature attribution e...
10/19/2021

Test time Adaptation through Perturbation Robustness

Data samples generated by several real world processes are dynamic in na...
05/09/2021

Truly shift-equivariant convolutional neural networks with adaptive polyphase upsampling

Convolutional neural networks lack shift equivariance due to the presenc...
01/21/2021

GhostSR: Learning Ghost Features for Efficient Image Super-Resolution

Modern single image super-resolution (SISR) system based on convolutiona...