3 papers
cs.IT2021
Mutual Information of Neural Network Initialisations: Mean Field Approximations
Jared Tanner, Giuseppe Ughi
The ability to train randomly initialised deep neural networks is known to depend strongly on the variance of the weight matrices and biases as well as the choice of nonlinear acti…
cs.LG2020
An Empirical Study of Derivative-Free-Optimization Algorithms for Targeted Black-Box Attacks in Deep Neural Networks
Giuseppe Ughi, Vinayak Abrol, Jared Tanner
We perform a comprehensive study on the performance of derivative free optimization (DFO) algorithms for the generation of targeted black-box adversarial attacks on Deep Neural Net…
cs.LG2020
A Model-Based Derivative-Free Approach to Black-Box Adversarial Examples: BOBYQA
Giuseppe Ughi, Vinayak Abrol, Jared Tanner
We demonstrate that model-based derivative free optimisation algorithms can generate adversarial targeted misclassification of deep networks using fewer network queries than non-mo…