LCANets++: Robust Audio Classification using Multi-layer Neural Networks with Lateral Competition

arXiv:2308.12882v2 Announce Type: replace-cross
Abstract: Audio classification aims at recognizing audio signals, including speech commands or sound events. However, current audio classifiers are susceptible to perturbations and adversarial attacks. In addition, real-world audio classification tasks often suffer from limited labeled data. To help bridge these gaps, previous work developed neuro-inspired convolutional neural networks (CNNs) with sparse coding via the Locally Competitive Algorithm (LCA) in the first layer (i.e., LCANets) for computer vision. LCANets learn in a combination of supervised and …

abstract adversarial adversarial attacks arxiv attacks audio bridge classification classifiers competition cs.cr cs.lg cs.sd current data eess.as events however layer networks neural networks robust sound speech tasks type work world