Sacho
@potosacho.bsky.social
PhD student in machine learning with Francis Bach &
Michael I. Jordan: uncertainty quantification, learning theory.
https://elsacho.github.io
Michael I. Jordan: uncertainty quantification, learning theory.
https://elsacho.github.io
Finally, clear proof and explanation of why overfitting a dataset leads to better accuracy: the refinement keeps improving—it’s just the calibration error that increases! Inspiring and a great breakthrough!
February 3, 2025 at 1:07 PM
Finally, clear proof and explanation of why overfitting a dataset leads to better accuracy: the refinement keeps improving—it’s just the calibration error that increases! Inspiring and a great breakthrough!