Luke Sanford
@lcsanford.bsky.social
Assistant Prof. at Yale School of the Environment. Political economy of climate and environment, land use change, remote sensing, causal ML. https://sanford-lab.github.io/
Here's what that image was supposed to look like:
March 18, 2025 at 12:12 AM
Here's what that image was supposed to look like:
We went for roads since it's easy to see how that measurement error could arise. Often we have no idea why RS + ML errors occur
March 18, 2025 at 12:10 AM
We went for roads since it's easy to see how that measurement error could arise. Often we have no idea why RS + ML errors occur
6/9
We then use a labeled forest cover data from high-resolution imagery. When comparing the ML predictions to ground-truth labels, a naive model under-estimates forest cover near roads. Our adversarial model, by contrast, recovers unbiased estimates, giving more reliable coefficients.
We then use a labeled forest cover data from high-resolution imagery. When comparing the ML predictions to ground-truth labels, a naive model under-estimates forest cover near roads. Our adversarial model, by contrast, recovers unbiased estimates, giving more reliable coefficients.
March 17, 2025 at 7:30 PM
6/9
We then use a labeled forest cover data from high-resolution imagery. When comparing the ML predictions to ground-truth labels, a naive model under-estimates forest cover near roads. Our adversarial model, by contrast, recovers unbiased estimates, giving more reliable coefficients.
We then use a labeled forest cover data from high-resolution imagery. When comparing the ML predictions to ground-truth labels, a naive model under-estimates forest cover near roads. Our adversarial model, by contrast, recovers unbiased estimates, giving more reliable coefficients.
5/9
We induce measurement error bias in a simulation of the effect of roads on forest cover. We show that a naive model yields biased estimates of this relationship, while an adversarial model gets it right.
We induce measurement error bias in a simulation of the effect of roads on forest cover. We show that a naive model yields biased estimates of this relationship, while an adversarial model gets it right.
March 17, 2025 at 7:30 PM
5/9
We induce measurement error bias in a simulation of the effect of roads on forest cover. We show that a naive model yields biased estimates of this relationship, while an adversarial model gets it right.
We induce measurement error bias in a simulation of the effect of roads on forest cover. We show that a naive model yields biased estimates of this relationship, while an adversarial model gets it right.
4/9
We also introduce a simple bias test: regress the ML prediction errors on your independent variable. If nonzero, you have measurement error bias. If you run that test while gathering ground-truth data, you can estimate how many labeled observations you’ll need to reject a target amount of bias.
We also introduce a simple bias test: regress the ML prediction errors on your independent variable. If nonzero, you have measurement error bias. If you run that test while gathering ground-truth data, you can estimate how many labeled observations you’ll need to reject a target amount of bias.
March 17, 2025 at 7:30 PM
4/9
We also introduce a simple bias test: regress the ML prediction errors on your independent variable. If nonzero, you have measurement error bias. If you run that test while gathering ground-truth data, you can estimate how many labeled observations you’ll need to reject a target amount of bias.
We also introduce a simple bias test: regress the ML prediction errors on your independent variable. If nonzero, you have measurement error bias. If you run that test while gathering ground-truth data, you can estimate how many labeled observations you’ll need to reject a target amount of bias.
3/9
Here’s how: a primary model predicts the outcome, while an adversarial model tries to predict the treatment using the prediction errors. As the adversary learns how to predict treatment, the primary model learns to make predictions where the errors contain no information about the treatment.
Here’s how: a primary model predicts the outcome, while an adversarial model tries to predict the treatment using the prediction errors. As the adversary learns how to predict treatment, the primary model learns to make predictions where the errors contain no information about the treatment.
March 17, 2025 at 7:30 PM
3/9
Here’s how: a primary model predicts the outcome, while an adversarial model tries to predict the treatment using the prediction errors. As the adversary learns how to predict treatment, the primary model learns to make predictions where the errors contain no information about the treatment.
Here’s how: a primary model predicts the outcome, while an adversarial model tries to predict the treatment using the prediction errors. As the adversary learns how to predict treatment, the primary model learns to make predictions where the errors contain no information about the treatment.
Still:
From www.theguardian.com/us-news/2025...
From www.theguardian.com/us-news/2025...
March 8, 2025 at 9:41 AM
Still:
From www.theguardian.com/us-news/2025...
From www.theguardian.com/us-news/2025...
Perfect Christmas Eve: ice skating in the snow on the neighborhood pond
December 24, 2024 at 5:46 PM
Perfect Christmas Eve: ice skating in the snow on the neighborhood pond
November 28, 2024 at 9:35 PM
Testing some new winter designs #poliscibakes
November 16, 2024 at 3:01 PM
Testing some new winter designs #poliscibakes
Chat GPT prompt: “Based on what you know about me make an image about what you think my life looks like” 🤨
November 16, 2024 at 12:51 PM
Chat GPT prompt: “Based on what you know about me make an image about what you think my life looks like” 🤨
Back to working only when the (new) tiny human allows it
November 13, 2024 at 11:47 AM
Back to working only when the (new) tiny human allows it
Total solar eclipse exceeded my high expectations. So cool.
April 9, 2024 at 1:46 PM
Total solar eclipse exceeded my high expectations. So cool.
This is a DiD meme. Are those…. bees? On his suit?
December 5, 2023 at 3:23 PM
This is a DiD meme. Are those…. bees? On his suit?
I asked GPT4 to give me a twfe meme image and this is what I got. @khoavuumn.bsky.social looks like your spot atop the econometrics meme podium is safe for now
December 5, 2023 at 1:44 PM
I asked GPT4 to give me a twfe meme image and this is what I got. @khoavuumn.bsky.social looks like your spot atop the econometrics meme podium is safe for now
More thanksgiving sourdough loaves
#poliscibakes
#poliscibakes
November 27, 2023 at 12:53 AM
More thanksgiving sourdough loaves
#poliscibakes
#poliscibakes
November 19, 2023 at 8:31 PM
