Change surfaces for expressive multidimensional changepoints and counterfactual prediction

William Herlands, Daniel B. Neill, Hannes Nickisch, Andrew Gordon Wilson

Research output: Contribution to journalArticlepeer-review

Abstract

Identifying changes in model parameters is fundamental in machine learning and statistics. However, standard changepoint models are limited in expressiveness, often addressing unidimensional problems and assuming instantaneous changes. We introduce change surfaces as a multidimensional and highly expressive generalization of changepoints. We provide a model-agnostic formalization of change surfaces, illustrating how they can provide variable, heterogeneous, and non-monotonic rates of change across multiple dimensions. Additionally, we show how change surfaces can be used for counterfactual prediction. As a concrete instantiation of the change surface framework, we develop Gaussian Process Change Surfaces (GPCS). We demonstrate counterfactual prediction with Bayesian posterior mean and credible sets, as well as massive scalability by introducing novel methods for additive nonseparable kernels. Using two large spatio-temporal datasets we employ GPCS to discover and characterize complex changes that can provide scientific and policy relevant insights. Specifically, we analyze twentieth century measles incidence across the United States and discover previously unknown heterogeneous changes after the introduction of the measles vaccine. Additionally, we apply the model to requests for lead testing kits in New York City, discovering distinct spatial and demographic patterns.

Original languageEnglish (US)
JournalJournal of Machine Learning Research
Volume20
StatePublished - Jun 1 2019

Keywords

  • Change surface
  • Changepoint
  • Counterfactual
  • Gaussian process
  • Kernel method
  • Scalable inference

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Statistics and Probability
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Change surfaces for expressive multidimensional changepoints and counterfactual prediction'. Together they form a unique fingerprint.

Cite this