Defending against adversarial attacks by randomized diversification
This addresses security concerns for ML applications by providing a defense mechanism against adversarial attacks, though it appears incremental as it builds on existing randomization and multi-channel concepts.
The paper tackles the vulnerability of machine learning systems to adversarial attacks by proposing a randomized diversification defense strategy, which uses a multi-channel architecture with secret key-based randomization to increase robustness, as shown in experimental evaluations against state-of-the-art attacks.
The vulnerability of machine learning systems to adversarial attacks questions their usage in many applications. In this paper, we propose a randomized diversification as a defense strategy. We introduce a multi-channel architecture in a gray-box scenario, which assumes that the architecture of the classifier and the training data set are known to the attacker. The attacker does not only have access to a secret key and to the internal states of the system at the test time. The defender processes an input in multiple channels. Each channel introduces its own randomization in a special transform domain based on a secret key shared between the training and testing stages. Such a transform based randomization with a shared key preserves the gradients in key-defined sub-spaces for the defender but it prevents gradient back propagation and the creation of various bypass systems for the attacker. An additional benefit of multi-channel randomization is the aggregation that fuses soft-outputs from all channels, thus increasing the reliability of the final score. The sharing of a secret key creates an information advantage to the defender. Experimental evaluation demonstrates an increased robustness of the proposed method to a number of known state-of-the-art attacks.