Generating Ensemble-Based Adversarial Attacks and Defense Methods for Robust Online Fraud Detection
Get an Alertwhen The Register has something new about ...
Check the boxes & select Email or Atom/RSS Feed.
With the evolving and diverse nature of fraud patterns in online payment systems, it's important to optimize explicitly for recognition and defense against such attacks. NVIDIA will introduce a cohort of domain-oriented adversarial attacks based on an ensemble of attack vector generation techniques, to optimize for discovery of sparse perturbation spaces yielding decisive changes in fraud adjudication. They will demonstrate the effectiveness of such ensembles in surpassing the attack capabilities of isolated attack frameworks. To proactively neutralize ensemble-oriented attacks, generative defensive procedures using adversarial retraining algorithms will be showcased. They will also address key challenges and applications in effective deployment of such models in large-scale fraud detection systems.
Nitin Sharma, Distinguished Scientist, AI Research Group, PayPal, Inc.
Please register, or log in to your account, to access this resource.