Exploiting the relationship between Kendall’s rank correlation and cosine similarity for attribution protection

• Main Authors: Wang, Fan, Kong, Adams Wai Kin
• Other Authors: School of Computer Science and Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2022
• Subjects: Computer Science - Learning; Computer Science - Artificial Intelligence; Artificial Intelligence; Neural Networks
• Online Access: https://hdl.handle.net/10356/161935; https://nips.cc/
• Institution: Nanyang Technological University

Model attributions are important in deep neural networks as they aid practitioners in understanding the models, but recent studies reveal that attributions can be easily perturbed by adding imperceptible noise to the input. The non-differentiable Kendall's rank correlation is a key performan...