Skip to content
Carmelics
Topics
Thinkers
Changes
Contributors
Loading account…
Statements
321,452
Perspectives
108,905
Topics
42
Home
/
Original
/
inverse
See Original
Inverse View
It is not the case that The standard bias-variance decomposition does not hold under all loss functions.
?
Set your confidence on the premises below to see your aggregate.
Reasons For
1 perspective
Reason for
?
1.
The bias-variance decomposition is derived specifically from squared loss.
?
How convincing is this?
Think about whether this reason is strong or weak
2.
Under a 0-1 loss function, all non-zero errors are treated equally regardless of magnitude.
?
How convincing is this?
Think about whether this reason is strong or weak
3.
Under 0-1 loss, bias and variance combine multiplicatively rather than additively.
?
How convincing is this?
Think about whether this reason is strong or weak
Next step
Based on where you are in your exploration
Browse more in Truth & Knowledge
Related propositions within the same area of thought.