Latent-space perturbations of VAE-encoded business process traces generate realistic-looking adversarial examples that flip outcome predictions more often than input-space permutations.
In: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generating Realistic Adversarial Examples for Business Processes using Variational Autoencoders
Latent-space perturbations of VAE-encoded business process traces generate realistic-looking adversarial examples that flip outcome predictions more often than input-space permutations.