Decreasing Model Stealing Querying for Black Box Adversarial Attacks

Black-box Adversarial Attacks using Substitute models: Effects of Data Distributions on Sample Transferability

A deep dive into the robustness of AdaBoost Ensembling combined with Adversarial Training

Effect of parameter tuning on reducing the number of queries required to perform model stealing

An Empirical Look at Gradient-based Black-box Adversarial Attacks on Deep Neural Networks Using One-point Residual Estimates