Tim Grube

TL;DR: The opportunities for more flexible decentralized control over the training process and its impact on the participants' privacy are introduced and the possibilities to enhance the efficiency and effectiveness of defense and verification methods are investigated.

TL;DR: This article reviews basic security and privacy requirements and introduces a new attacker model based on statistical disclosure, used to challenge anonymity, and presents a novel system that leverages peer-to-peer networking concepts; this novel approach protects subscriber anonymity by means of Probabilistic Forwarding and through a novel so-called Shell Game algorithm.

TL;DR: This work proposes and compares sampling strategies to sample the data based on social graphs in a privacy-preserving manner and shows that the sampling-based approach can achieve a high average precision score with as low as 50% sampling rate and, therefore, with a 50% reduction of communication cost.

TL;DR: A parameter study is performed to analyze the influence of varying participation, utilization, and timing properties on anonymity and efficiency and indicates that cover traffic can be generated more efficiently while anonymity is still on a reasonable level.