Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness. (arXiv:2206.07839v1 [cs.LG])

Web: http://arxiv.org/abs/2206.07839

Certifiable robustness is a highly desirable property for adopting deep
neural networks (DNNs) in safety-critical scenarios, but often demands tedious
computations to establish. The main hurdle lies in the massive amount of
non-linearity in large DNNs. To trade off the DNN expressiveness (which calls
for more non-linearity) and robustness certification scalability (which prefers
more linearity), we propose a novel solution to strategically manipulate
neurons, by "grafting" appropriate levels of linearity. The core of our
proposal is to first linearize insignificant …

arxiv lg pruning robustness