If an ML system depends on abilities that _____ cannot reliably perform, some standard error-analysis techniques stop being useful.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Why is error analysis especially useful when the machine learning task is something people can already do reliably?
When people can perform a task reliably, their performance can be used as a reference point for analyzing model errors.
Human-Solvable Tasks and Better _____
Match each task situation to the error-analysis implication it supports.
Using Human Performance as a Benchmark for Error Analysis
What should a team expect when its system solves a problem that experienced people cannot judge reliably?
Error Analysis Always Works the Same Way for Every Task
Match each machine learning scenario to its consequence for error analysis tools.
Order the reasoning steps showing how human-performable tasks support better project prioritization.
Using human performance as a reference in machine learning work
When Human Benchmarks Are Unavailable
Why some error analysis methods need a human baseline