Black-box statistics, model testing, and evaluation (Chatbot Arena).
@Berkeley_EECS Ph.D. and student researcher @GoogleDeepMind. Former @stanford_ee.
Aug 1, 2023 • 7 tweets • 2 min read
🤖Conformal PID Control for Time-Series Prediction🕹
📝w/Candès & Tibshirani!
Is conformal prediction out of control🤪?
Yes! CP can be thought of as PID control, giving stronger guarantees+algorithms that predict distribution shifts as they happen! A🧵 https://t.co/3NvH41Gw2farxiv.org/abs/2307.16895
Before I explain how it works, check this result on COVID death forecasting.
The official CDC forecast missed the upswing. Running PID control on top improved the result — by a lot.
Why? Because the errors were predictable, so it just PREDICTED them! Hence, better coverage :)
Jan 25, 2023 • 5 tweets • 2 min read
📯Prediction-Powered Inference📯
With the rise of AlphaFold etc., people are using ML predictions to replace costly experimental data.
But predictions aren't perfect; can we still use them for rigorous downstream inferences?
The answer: yes. A 🧵 arxiv.org/abs/2301.09633
The key idea is to define a mathematical object called
Δ, the rectifier.
We use Δ to "rectify" the predictions in a way that recovers the ground truth estimand.
For means, Δ is a bias, but in general, it is not. We can use Δ to form a prediction-powered confidence set (green).
Aug 5, 2022 • 5 tweets • 3 min read
I’m thrilled to announce Conformal Risk Control: a way to bound quantities other than coverage with conformal prediction.