Reproduce “EarthquakeNPP: A Benchmark for Earthquake Forecasting with Neural Point Processes” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Samuel Stockman, Daniel Lawson, Maximilian Werner
The authors introduce EarthquakeNPP, a chronological benchmark combining California earthquake catalogs, ETAS, neural point processes, and seismology’s evaluation procedures. It enables forecasts to update continuously and compares likelihood and generative tests. Across five neural implementations, ETAS generally performs better, including a 95.9% versus 68.0% versus 0.8% SCEDC pseudo-likelihood pass rate, but results vary by test and dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.