Alex Luedtke
@alexluedtke.bsky.social
statistician at harvard med school • causal inference, machine learning, nonparametrics
alexluedtke.com
alexluedtke.com
New paper on generative modeling of counterfactual distributions! We give a way to answer "what if" questions with generative models.
For example: what would faces look like if they were all smiling?
arxiv.org/abs/2509.16842
For example: what would faces look like if they were all smiling?
arxiv.org/abs/2509.16842
September 24, 2025 at 8:42 PM
New paper on generative modeling of counterfactual distributions! We give a way to answer "what if" questions with generative models.
For example: what would faces look like if they were all smiling?
arxiv.org/abs/2509.16842
For example: what would faces look like if they were all smiling?
arxiv.org/abs/2509.16842
New paper, led by my student Alex Kokot!
We study dataset compression through coreset selection - finding a small, weighted subset of observations that preserves information with respect to some divergence.
arxiv.org/abs/2504.20194
We study dataset compression through coreset selection - finding a small, weighted subset of observations that preserves information with respect to some divergence.
arxiv.org/abs/2504.20194
April 30, 2025 at 12:59 PM
New paper, led by my student Alex Kokot!
We study dataset compression through coreset selection - finding a small, weighted subset of observations that preserves information with respect to some divergence.
arxiv.org/abs/2504.20194
We study dataset compression through coreset selection - finding a small, weighted subset of observations that preserves information with respect to some divergence.
arxiv.org/abs/2504.20194
New paper!
arxiv.org/abs/2405.08675
tldr: automatic differentiation can be used to derive efficient influence functions and construct efficient estimators.
arxiv.org/abs/2405.08675
tldr: automatic differentiation can be used to derive efficient influence functions and construct efficient estimators.
June 25, 2024 at 4:36 PM
New paper!
arxiv.org/abs/2405.08675
tldr: automatic differentiation can be used to derive efficient influence functions and construct efficient estimators.
arxiv.org/abs/2405.08675
tldr: automatic differentiation can be used to derive efficient influence functions and construct efficient estimators.