Kernelized Capsule Networks
This work addresses robustness and uncertainty issues in Capsule Networks for image recognition, offering an incremental improvement with potential applications in adversarial defense.
The paper tackled improving the robustness and uncertainty estimation of Capsule Networks by integrating non-parametric kernel methods, achieving comparable prediction performance while enhancing adversarial robustness and providing uncertainty measures for adversarial detection.
Capsule Networks attempt to represent patterns in images in a way that preserves hierarchical spatial relationships. Additionally, research has demonstrated that these techniques may be robust against adversarial perturbations. We present an improvement to training capsule networks with added robustness via non-parametric kernel methods. The representations learned through the capsule network are used to construct covariance kernels for Gaussian processes (GPs). We demonstrate that this approach achieves comparable prediction performance to Capsule Networks while improving robustness to adversarial perturbations and providing a meaningful measure of uncertainty that may aid in the detection of adversarial inputs.