Boosting Adversarial Robustness using Feature Level Stochastic Smoothing
This work addresses the need for higher robust accuracy and confidence assignment in critical domains like robotics and autonomous navigation, representing an incremental improvement over current adversarial defenses.
The paper tackles the problem of insufficient adversarial robustness in deep neural networks for critical applications by proposing a generic method that introduces stochasticity to smooth decision boundaries and reject low-confidence predictions, resulting in boosted robustness on accepted samples and improved robustness without rejection over existing adversarial training methods.
Advances in adversarial defenses have led to a significant improvement in the robustness of Deep Neural Networks. However, the robust accuracy of present state-ofthe-art defenses is far from the requirements in critical applications such as robotics and autonomous navigation systems. Further, in practical use cases, network prediction alone might not suffice, and assignment of a confidence value for the prediction can prove crucial. In this work, we propose a generic method for introducing stochasticity in the network predictions, and utilize this for smoothing decision boundaries and rejecting low confidence predictions, thereby boosting the robustness on accepted samples. The proposed Feature Level Stochastic Smoothing based classification also results in a boost in robustness without rejection over existing adversarial training methods. Finally, we combine the proposed method with adversarial detection methods, to achieve the benefits of both approaches.