A Causal View on Robustness of Neural Networks

05/03/2020
by   Cheng Zhang, et al.
22

We present a causal view on the robustness of neural networks against input manipulations, which applies not only to traditional classification tasks but also to general measurement data. Based on this view, we design a deep causal manipulation augmented model (deep CAMA) which explicitly models possible manipulations on certain causes leading to changes in the observed effect. We further develop data augmentation and test-time fine-tuning methods to improve deep CAMA's robustness. When compared with discriminative deep neural networks, our proposed model shows superior robustness against unseen manipulations. As a by-product, our model achieves disentangled representation which separates the representation of manipulations from those of other latent causes.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 6

10/22/2020

Learning Loss for Test-Time Augmentation

Data augmentation has been actively studied for robust neural networks. ...
06/03/2019

Achieving Generalizable Robustness of Deep Neural Networks by Stability Training

We study the recently introduced stability training as a general-purpose...
02/12/2020

Efficient Training of Deep Convolutional Neural Networks by Augmentation in Embedding Space

Recent advances in the field of artificial intelligence have been made p...
10/05/2021

Distribution Mismatch Correction for Improved Robustness in Deep Neural Networks

Deep neural networks rely heavily on normalization methods to improve th...
07/11/2012

Robustness of Causal Claims

A causal claim is any assertion that invokes causal relationships betwee...
12/01/2021

Inducing Causal Structure for Interpretable Neural Networks

In many areas, we have well-founded insights about causal structure that...
12/01/2021

Robustness in Deep Learning for Computer Vision: Mind the gap?

Deep neural networks for computer vision tasks are deployed in increasin...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.