Elliot Creager

Assistant Professor at University of Waterloo. Faculty Affiliate at Vector Institute for Artificial Intelligence and Schwartz Reisman Institute for Technology and Society.


Welcome to my humble homepage!

I conduct research on how machine learning methods fail, and what can be done to improve their robustness in new contexts. My methods draw on technical approaches such as representation learning, causal inference, and data augmentation to build models for prediction and control that generalize beyond the available training data.

Meanwhile, I am very interested in analyzing the broader implications of using machine learning to automate decisions, especially the tendency of this process to codify and compound existing social inequities.

Outside of work, I enjoy birdwatching around the Greater Toronto Area, where I’ve seen ~120 species. Also, I am a guitarist in the Adam Solmon Band.


Sep 1, 2023 :pencil: I have joined University of Waterloo as an Assistant Professor.
Aug 21, 2023 :mortar_board: After passing my final oral examination (external thesis defense), I have now completed my PhD!
Jul 19, 2023 :soon: I will be joining University of Waterloo’s Department of Electrical and Computer Engineering as an Assistant Professor this fall!
Jul 13, 2023 :ocean: SurfsUp was accepted to ICCV 2023!
Jul 10, 2023 :lock: I gave an invited talk at Vector Institute’s Machine Learning Security and Privacy Workshop.
Jun 22, 2023 :notebook: I attended the SRI’s Absolutely Interdisciplinary conference in Toronto.
Jun 19, 2023 :boxing_glove: “Out of the Ordinary: Spectrally Adapting Regression for Covariate Shift” (preprint forthcoming) was accepted to the ICML 2023 Workshop on Spurious Correlations, Invariance, and Stability!
Jun 16, 2023 :hotdog: I attended FAccT 2023 in Chicago.
May 31, 2023 :book: I passed my department thesis examination, where I successfully defended my thesis “Robust Machine Learning by Transforming and Augmenting Imperfect Training Data”.
Apr 27, 2023 :microphone: I gave an invited talk at the Mila reading group DEFirst, which studies fairness and explainability in information retrieval.
Apr 24, 2023 :pencil2: I served as panelist for an event on “Generative AI and the future of education” hosted by SRI and OISE.
Apr 13, 2023 :ocean: SurfsUp, a neural fluid simulator that can generalize OOD by simulating how fluids interact with novel surfaces, is available as a preprint.
Feb 25, 2023 :microscope: I discussed the ethics of AI research with Grade 12 students as a panelist for Pursue STEM.
Feb 2, 2023 :statue_of_liberty: I attended the annual Algorithmic Fairness meeting at the Simons Institute in NYC
Dec 5, 2022 :microphone: I co-presented the tutorial Algorithmic Fairness: at the Intersections at NeurIPS 2022
Nov 17, 2022 :microphone: I gave an invited talk (“Methods for Counterfactual Data Augmentation in RL”) at Vector Institute.
Nov 7, 2022 :tophat: I was named as a “Top Reviewer” of NeurIPS 2022 papers
Nov 6, 2022 :mag: I am now on the job market! Searching for faculty, industry and post-doc positions…
Sep 14, 2022 :tada: “MoCoDA: Model-based Counterfactual Data Augmentation” was accepted to NeurIPS 2022
Sep 2, 2022 :pencil: I will be co-presenting the NeurIPS 2022 tutorial “Algorithmic Fairness: at the Intersections” alongside Golnoosh Farnadi and Q. Vera Liao

selected publications

  1. ICML
    Environment Inference for Invariant Learning
    Elliot Creager, Joern-Henrik Jacobsen, and Richard Zemel
    In Proceedings of the 38th International Conference on Machine Learning Jul 2021
  2. NeurIPS
    Counterfactual Data Augmentation using Locally Factored Dynamics
    Silviu Pitis, Elliot Creager, and Animesh Garg
    In Advances in Neural Information Processing Systems Dec 2020
  3. ICML
    Flexibly Fair Representation Learning by Disentanglement
    Elliot Creager, David Madras, Joern-Henrik Jacobsen, Marissa Weis, Kevin Swersky, Toniann Pitassi, and Richard Zemel
    In Proceedings of the 36th International Conference on Machine Learning Jun 2019