Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 289 - Recent Advances in Mathematical Statistics and Probability
Type: Contributed
Date/Time: Wednesday, August 11, 2021 : 1:30 PM to 3:20 PM
Sponsor: IMS
Abstract #317876
Title: Adversarial Contamination of Networks: A New Trimming Method
Author(s): Sheyda Peyman* and Vince Lyzinski
Companies: University of Maryland and University of Maryland
Keywords: vertex nomination; spectral embedding; adversarial contamination; robustness; networks; clustering
Abstract:

As graph data becomes more ubiquitous, the need for robust inferential graph algorithms to operate in these complex data domains is crucial. In many cases of interest, inference is further complicated by the presence of adversarial data contamination. The adversary frequently changes the data distribution in ways that negatively affect statistical inference and algorithmic performance. We study this phenomenon in the context of vertex nomination, a semi-supervised information retrieval task for network data. In vertex nomination, a common suite of methods relies on spectral graph embeddings, which have been shown to provide both good algorithmic performance and a flexible setting in which regularization techniques can be implemented to help mitigate the effect of an adversary. Many current regularization methods rely on direct network trimming to effectively excise the adversarial contamination, although this direct trimming often gives rise to complicated dependency structures in the resulting graph. We propose a new trimming method that operates in model space, which is more amenable to theoretical analysis and demonstrates superior performance in a number of simulations.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program