Quantum-inspired analysis of neural network vulnerabilities: the role of conjugate variables in system attacks
Sign Up to like & getrecommendations! Published in 2024 at "National Science Review"
DOI: 10.1093/nsr/nwae141
Abstract: ABSTRACT Neural networks demonstrate 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… read more here.
Keywords: inspired analysis; network vulnerabilities; network; analysis neural ... See more keywords
Bio-inspired Analysis of Deep Learning on Not-So-Big Data Using Data-Prototypes
Sign Up to like & getrecommendations! Published in 2019 at "Frontiers in Computational Neuroscience"
DOI: 10.3389/fncom.2018.00100
Abstract: Deep artificial neural networks are feed-forward architectures capable of very impressive performances in diverse domains. Indeed stacking multiple layers allows a hierarchical composition of local functions, providing efficient compact mappings. Compared to the brain, however,… read more here.
Keywords: big data; learning; bio inspired; deep learning ... See more keywords