TIMERS: Error-Bounded SVD Restart on Dynamic Networks

Understand

Singular Value Decomposition (SVD) is a popular approach in various network applications, such as link prediction and network parameter characterization.

• Incremental SVD approaches are proposed to process newly changed nodes and edges in dynamic networks.
• However, incremental SVD approaches suffer from serious error accumulation inevitably due to approximation on incremental updates.
• SVD restart is an effective approach to reset the aggregated error, but when to restart SVD for dynamic networks is not addressed in literature.