Go Back
You can't go any further
You can't go any further
meritocratic
regular
democratic
hot
top
alive
22 posts
Statistics
3 posts
Optimizer's curse
209
Reality is often underpowered
Gregory Lewis
3y
17
204
List of ways in which cost-effectiveness estimates can be misleading
saulius
3y
30
66
Probability distributions of Cost-Effectiveness can be misleading
Lorenzo Buonanno
5mo
52
39
Take care with notation for uncertain quantities
Gregory Lewis
2y
2
37
[Stats4EA] Expectations are not Outcomes
matthewp
2y
5
34
Estimating value from pairwise comparisons
Jonas Moss
2mo
3
32
Five Ways to Handle Flow-Through Effects
Peter Wildeford
6y
3
30
A peek at pairwise preference estimation in economics, marketing, and statistics
Jonas Moss
2mo
5
25
Turning percentages back into people: personalizing quantification
sky
2y
0
25
[Stats4EA] Uncertain Probabilities
matthewp
2y
7
22
How accurately does anyone know the global distribution of income?
Robert_Wiblin
5y
30
17
How to Understand and Mitigate Risk (Crosspost from LessWrong)
Halffull
3y
7
16
Statistics for Lazy People, Part 2
Hmash
1y
2
15
Expected value estimates you can take (somewhat) literally
Gregory Lewis
8y
24
84
[Link] The Optimizer's Curse & Wrong-Way Reductions
Chris Smith
3y
61
52
Don't Over-Optimize Things
Owen Cotton-Barratt
6mo
7
23
The Multi-Class Optimizer's Curse
FeelingSentient
1y
0