Accelerated Nonparametrics for Cascades of Poisson Processes

TL;DR

This paper introduces a computationally efficient nonparametric estimation method for cascades of Poisson processes, enabling analysis of large datasets by reducing complexity from quadratic to near-linear.

Contribution

It develops a novel nonparametric estimator assuming local triggering, with automatic domain learning and minimal hyperparameters, suitable for large-scale data.

Findings

Achieves $ ext{O}(N ext{log} N)$ computational complexity

Reduces storage complexity to $ ext{O}(N)$

Successfully applied to large seismic datasets

Abstract

Cascades of Poisson processes are probabilistic models for spatio-temporal phenomena in which (i) previous events may trigger subsequent events, and (ii) both the background and triggering processes are conditionally Poisson. Such phenomena are typically "data rich but knowledge poor", in the sense that large datasets are available yet a mechanistic understanding of the background and triggering processes which generate the data are unavailable. In these settings nonparametric estimation plays a central role. However existing nonparametric estimators have computational and storage complexity $\mathcal{O}(N^2)$, precluding their application on large datasets. Here, by assuming the triggering process acts only locally, we derive nonparametric estimators with computational complexity $\mathcal{O}(N\log N)$ and storage complexity $\mathcal{O}(N)$. Our approach automatically learns the domain of the triggering process from data and is essentially free from hyperparameters. The methodology is applied to a large seismic dataset where estimation under existing algorithms would be infeasible.