Maggie Makar

Assistant professor in CSE, University of Michigan E-mail: mmakar [at] umich [dot] edu Physical location: BBB 3769 Twitter: @Maggiemakar Mastodon: mmakar@mastodon.social CV: As of Sept 2022 Google scholar profile

About me

I study machine learning, causal inference, and the intersection between the two. My work focuses on building data-efficient causal inference methods in resource-constrained settings, and building robust predictive ML models using ideas from causality. Prior to UM, I completed my PhD in CSAIL, MIT and my bachelors degree in math and economics at Umass Amherst.

Joining my group

I am currently looking for PhD students to work with me. To get a better sense of my work and what my previous student collaborators have achieved please see the students page and the publications page.

Please read how to join before emailing me.

Selected Publications

Full list of publications here

• Hypothesis Testing the Circuit Hypothesis in LLMs
  C Shi, N Beltran-Velez, A Nazaret, C Zheng, A Garriga-Alonso, A Jesson, M Makar, D Blei
  NeurIPS 2024 [workshop version, full paper forthcoming]

• Offline policy evaluation and optimization under confounding
  C Kausik, Y Lu, K Tan, M Makar, Y Wang, A Tewari
  AIStats 2024 [paper]

• Partial identification of the maximum mean discrepancy with mismeasured data
  R. Nafshi, M. Makar
  UAI, 2024 [paper]

• Causally motivated multi-shortcut identification and removal
  J. Zheng, M. Makar
  NeurIPS 2022 [paper]