Deep Learning Models for Principled Causal Forecasting

Numerous Machine Learning (ML) tasks are forecasting problems, used to make downstream decisions. Acting on ML forecasts however can changes the distribution of observations relevant to the forecast. The implication is downstream decision optimization procedures implicitly expect the ML model to generalize outside of the observational distribution. Unfortunately, this is often not the case, and ML models tend to be brittle outside of their training distribution. ML models will thus produce unreliable extrapolations, leading to poor downstream decisions based on wrong forecasts.

Causal models, which aim to learn the structural causal models underlying the data generation process, are a natural fit for such use-cases. This is because causal mechanisms are more robust to superficial changes in the data distribution, and can be expected to extrapolate better to new environments. This project aims to combine deep learning and causal inference to develop causal forecasting models adapted to two important applications. Students will start by implementing existing approaches on one of three applications, before working on improvements to core causal modeling techniques.

Faculty Supervisor:

Mathias Lécuyer

Student:

Partner:

Institut Polytechnique de Paris

Discipline:

Computer science

Sector:

Education

University:

The University of British Columbia

Program: