<ul><li>Distributed learning (DL) uses multiple nodes to accelerate training, enabling efficient optimization of large-scale models.</li><li>Stochastic Gradient Descent (SGD) is a key optimization algorithm in DL but communication bottlenecks limit scalability and efficiency.</li><li>Compressed SGD techniques are used to address communication overheads, but they introduce trustworthiness concerns, including attacks like gradient inversion and membership inference attacks.</li><li>Empirical studies show that compressed SGD demonstrates higher resistance to privacy leakage compared to uncompressed SGD, and the reliability of using membership inference attacks as a metric for assessing privacy risks in distributed learning is questionable.</li></ul>

Trustworthiness of Stochastic Gradient Descent in Distributed Learning

Discover more