Manifold-aware Training: Increase Adversarial Robustness with Feature Clustering

no code implementations • 1 Jan 2021 • Ting-An Yen, Chun-Shien Lu, Pau-Choo Chung

Inspired by the observation from the distribution properties of the features extracted by the CNNs in the feature space and their link to robustness, this work designs a novel training process called Manifold-Aware Training (MAT), which forces CNNs to learn compact features to increase robustness.

Adversarial Robustness Clustering

NINEPINS: Nuclei Instance Segmentation with Point Annotations

no code implementations • 24 Jun 2020 • Ting-An Yen, Hung-Chun Hsu, Pushpak Pati, Maria Gabrani, Antonio Foncubierta-Rodríguez, Pau-Choo Chung

Deep learning-based methods are gaining traction in digital pathology, with an increasing number of publications and challenges that aim at easing the work of systematically and exhaustively analyzing tissue slides.

Instance Segmentation Pseudo Label +2