Timezone: »

 
Poster
ResNets Ensemble via the Feynman-Kac Formalism to Improve Natural and Robust Accuracies
Bao Wang · Zuoqiang Shi · Stanley Osher

Tue Dec 10 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #14
in Algorithms -- Adversarial Learning »
We unify the theory of optimal control of transport equations with the practice of training and testing of ResNets. Based on this unified viewpoint, we propose a simple yet effective ResNets ensemble algorithm to boost the accuracy of the robustly trained model on both clean and adversarial images. The proposed algorithm consists of two components: First, we modify the base ResNets by injecting a variance specified Gaussian noise to the output of each residual mapping. Second, we average over the production of multiple jointly trained modified ResNets to get the final prediction. These two steps give an approximation to the Feynman-Kac formula for representing the solution of a convection-diffusion equation. For the CIFAR10 benchmark, this simple algorithm leads to a robust model with a natural accuracy of {\bf 85.62}\% on clean images and a robust accuracy of ${\bf 57.94 \%}$ under the 20 iterations of the IFGSM attack, which outperforms the current state-of-the-art in defending against IFGSM attack on the CIFAR10.
Keywords: Deep Learning · Adversarial Networks · Deep Learning
[ Paper]  [ Poster] 

Author Information

Bao Wang (UCLA)
Zuoqiang Shi (zqshi@mail.tsinghua.edu.cn)
Stanley Osher (UCLA)

More from the Same Authors