Code for reproducing the ICML 2020 paper "Transfer Learning without Knowing: Reprogramming Black-box Machine Learning Models with Scarce Data and Limited Resources" synced with https://github.com/yunyuntsai/Black-box-Adversarial-Reprogramming
Our code is implemented in Python 3.6 and Tensorflow 1.14.
The following figure illustrates the framework for our proposed black-box adversarial reprogramming method (BAR):
-
Generate adversarial program.
-
Find q pertubed adversarial programs with vectors that are uniformly drawn at random from a unit Euclidean sphere.
-
Estimate gradient with zeroth-order gradient estimator. The corresponding algorithmic convergence guarantees have been proved in both the convex loss and non-convex loss settings (Liu et al., 2018; 2019).
-
Optimize adversarial program’s parameters W.
For more detail, please refer to our main paper, and video on slideslive!.