Concept icon
Concept

Why Human Baselines Are Less Helpful on Hard-to-Judge Tasks

When a task is one that people do poorly, such as ranking restaurant recommendations, forecasting weekly demand, or estimating loan default risk, a human-level benchmark is much less informative. In those settings, human labels may be unreliable, intuition is a weak guide, and it is hard to define a meaningful target error rate.

0

1

Concept icon
Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI