Certified Robustness via Randomized Smoothing over Multiplicative Parameters of Input Transformations
This work addresses certified robustness for image classifiers against specific multiplicative transformations, representing an incremental domain-specific advancement.
The paper tackles the problem of providing certified robustness against multiplicative gamma correction transformations by proposing a novel randomized smoothing approach over multiplicative parameters. The results show that using an asymmetrical Rayleigh distribution yields better robustness certificates than Gaussian, Laplace, or uniform distributions for certain perturbation parameters.
Currently the most popular method of providing robustness certificates is randomized smoothing where an input is smoothed via some probability distribution. We propose a novel approach to randomized smoothing over multiplicative parameters. Using this method we construct certifiably robust classifiers with respect to a gamma correction perturbation and compare the result with classifiers obtained via other smoothing distributions (Gaussian, Laplace, uniform). The experiments show that asymmetrical Rayleigh distribution allows to obtain better certificates for some values of perturbation parameters. To the best of our knowledge it is the first work concerning certified robustness against the multiplicative gamma correction transformation and the first to study effects of asymmetrical distributions in randomized smoothing.