Concept

Progress Slows After Machines Surpass Human-Level Performance

Once humans have a hard time identifying examples that an algorithm is clearly getting wrong, only a subset of human-comparison techniques still apply. Progress is therefore usually slower on problems where machines already surpass human-level performance, while it is faster when machines are still catching up to humans.

0

1

Updated 2026-05-25

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy