Additive Poisson Process: Learning Intensity of Higher-Order Interaction in Stochastic Processes

Simon Luo, Feng Zhou, Lamiae Azizi, Mahito Sugiyama

Keywords: Poisson Process, Log-Linear Model, Energy-Based Model, Generalized Additive Models, Information Geometry

Abstract: We present the Additive Poisson Process (APP), a novel framework that can model the higher-order interaction effects of the intensity functions in stochastic processes using lower dimensional projections. Our model combines the techniques in information geometry to model higher-order interactions on a statistical manifold and in generalized additive models to use lower-dimensional projections to overcome the effects from the curse of dimensionality. Our approach solves a convex optimization problem by minimizing the KL divergence from a sample distribution in lower dimensional projections to the distribution modeled by an intensity function in the stochastic process. Our empirical results show that our model is able to use samples observed in the lower dimensional space to estimate the higher-order intensity function with extremely sparse observations.

One-sentence Summary: Efficient estimation of high dimensional intensity function using information geometric projections.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=0ObjVkJ9oo