Skip to yearly menu bar Skip to main content


Invited Talk
in
Workshop: MATH-AI: Toward Human-Level Mathematical Reasoning

Has Progress on Math been Surprising?

Jacob Steinhardt


Abstract:

In 2021, we commissioned forecasters to predict progress on ML benchmarks, including the MATH dataset for mathematical problem-solving. Progress on MATH ended up being much faster than predicted. I'll discuss what we should and shouldn't take away from this, my own predictions for future progress, and general implications for predicting future developments in ML.

Chat is not available.