Deep Learning and Music Adversaries

Corey Kereliuk, Bob L. Sturm, Jan Larsen

AbstractAn adversary is essentially an algorithm intent on
making a classification system perform in some particular way
given an input, e.g., increase the probability of a false negative.
Recent work builds adversaries for deep learning systems applied
to image object recognition, which exploits the parameters of
the system to find the minimal perturbation of the input image
such that the network misclassifies it with high confidence. We
adapt this approach to construct and deploy an adversary of
deep learning systems applied to music content analysis. In our
case, however, the input to the systems is magnitude spectral
frames, which requires special care in order to produce valid
input audio signals from network-derived perturbations. For two
different train-test partitionings of two benchmark datasets, and
two different deep architectures, we find that this adversary is
very effective in defeating the resulting systems. We find the
convolutional networks are more robust, however, compared with
systems based on a majority vote over individually classified
audio frames. Furthermore, we integrate the adversary into the
training of new deep systems, but do not find that this improves
their resilience against the same adversary.
Keywordsdeep nural networks, music information retrieval, content based processing, pattern recognition and claasicificastion
TypeJournal paper [With referee]
JournalIEEE Transactions on Multimedia
Year2015    Month November
PublisherIEEE
ISBN / ISSNDOI 10.1109/TMM.2015.2478068
NoteAeepar in 'Deep Learning for Multimedia Computing' special section
Electronic version(s)[pdf]
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing


Back  ::  IMM Publications