Bayesian Nonparametric Poisson-Process Allocation for Time-Sequence Modeling

TL;DR

The paper introduces BaNPPA, a Bayesian nonparametric model for time-sequences that automatically determines the number of latent functions using Gaussian processes, with an efficient inference algorithm.

Contribution

It presents a novel nonparametric Poisson process model for time-sequences that infers latent functions and addresses unidentifiability issues during inference.

Findings

Efficient variational inference algorithm developed.

Model successfully infers latent functions on synthetic data.

Demonstrates scalability and effectiveness on real-world data.

Abstract

Analyzing the underlying structure of multiple time-sequences provides insights into the understanding of social networks and human activities. In this work, we present the \emph{Bayesian nonparametric Poisson process allocation} (BaNPPA), a latent-function model for time-sequences, which automatically infers the number of latent functions. We model the intensity of each sequence as an infinite mixture of latent functions, each of which is obtained using a function drawn from a Gaussian process. We show that a technical challenge for the inference of such mixture models is the unidentifiability of the weights of the latent functions. We propose to cope with the issue by regulating the volume of each latent function within a variational inference algorithm. Our algorithm is computationally efficient and scales well to large data sets. We demonstrate the usefulness of our proposed model through experiments on both synthetic and real-world data sets.