Adaptive Clustering of Robust Semantic Representations for Adversarial Image Purification

no code implementations • 5 Apr 2021 • Samuel Henrique Silva, Arun Das, Ian Scarff, Peyman Najafirad

In order to evaluate the most adequate SRD, we rely on the distance between robust latent representations and semantic cluster distributions.

Clustering Semantic Similarity +1

Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey

no code implementations • 1 Jul 2020 • Samuel Henrique Silva, Peyman Najafirad

We provide a taxonomy to classify adversarial attacks and defenses, formulate the Robust Optimization problem in a min-max setting and divide it into 3 subcategories, namely: Adversarial (re)Training, Regularization Approach, and Certified Defenses.

Adversarial Robustness BIG-bench Machine Learning