Learn Before
When to Repair Noisy Development Labels
Development-set labels should be corrected when the mislabeled fraction is large enough to interfere with fast algorithm comparison and make it hard to tell whether one candidate model is better than another.
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
Adding a Mislabeled Column to Error Review
When to Repair Noisy Development Labels
Keep Validation and Test Label Fixes Aligned
In dev-set error analysis, what does it mean for an example to be mislabeled?
True or False: Mislabeled examples in a dev set are mistakes made by the model.
In a labeled example written as (x, _____), the second component is the target label.
Match each symbol in a mislabeled training example to its meaning.
Arrange the events that lead to finding a mislabeled validation example in a text classifier project.
What makes a dev-set example mislabeled?
Find the data problem in a traffic-sign classifier's dev set.
Where does a mislabeled dev example get its error?
Which option shows a mislabeled example in a wildlife image dataset used to classify birds?
True or False: In a mislabeled training example (x, y), the value y is wrong.
Learn After
Why mislabeled development examples matter more as models improve
When should mislabeled development-set examples be fixed?
When to Repair Validation Labels
Noisy labels can ____ the comparison itself.
Match each development-set concept to its role in deciding whether labels should be fixed.
Order the reasoning for deciding whether a noisy development set should be corrected.
When should mislabeled dev-set examples be fixed?
Decide When Incorrect Validation Labels Should Be Fixed
When is it worth fixing mislabeled validation examples?
When is cleaning dev-set labels worth the effort?
Fixing validation labels one by one is always required.