Quantum-Inspired Analysis of Neural Network Vulnerabilities: The Role of Conjugate Variables in System Attacks (2402.10983v1)
Abstract: Neural networks demonstrate inherent vulnerability to small, non-random perturbations, emerging as adversarial attacks. Such attacks, born from the gradient of the loss function relative to the input, are discerned as input conjugates, revealing a systemic fragility within the network structure. Intriguingly, a mathematical congruence manifests between this mechanism and the quantum physics' uncertainty principle, casting light on a hitherto unanticipated interdisciplinarity. This inherent susceptibility within neural network systems is generally intrinsic, highlighting not only the innate vulnerability of these networks but also suggesting potential advancements in the interdisciplinary area for understanding these black-box networks.
- Heisenberg W. Über den anschaulichen inhalt der quantentheoretischen kinematik und mechanik. Zeitschrift für Physik 1927; 43: 172–198.
- Bohr N. On the notions of causality and complementarity. Science 1950; 111: 51–54.
- Agarwal A, Singh R and Vatsa M. The role of ’sign’ and ’direction’ of gradient on the performance of cnn. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (2020) 2748–2756.