Researchers have established the minimum observation time required to accurately reconstruct latent network structures from event-based data. For Hawkes processes with sparse, weak interactions among 'd' entities, an observation time proportional to log(d) is both sufficient and necessary. The study proposes a two-stage estimation method involving data screening and least-squares refinement for the upper bound, and employs Fano's inequality with Jacod's Girsanov formula for the lower bound. AI
IMPACT Provides theoretical underpinnings for understanding complex system dynamics, potentially impacting AI models that analyze event-based data.
RANK_REASON Academic paper detailing theoretical results on network inference. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →