skip to content
promptdojo
_
[ save your spot ]
[ follow on x ]
Dimensionality and distance traps — step 2 of 7
ch 40 · unsupervised learning, embeddings, and recommenders
2/7
promptdojo
_
›
phase 08 · ai/ml engineering
›
ch 40 · unsupervised learning, embeddings, and recommenders
lesson 2 of 4 · dimensionality and distance traps
step 2 of 7 in this lesson
What is the "curse of dimensionality" for distance-based methods like nearest-neighbor?
1
As dimensions grow, points become nearly equidistant, so "nearest" stops being meaningful.
2
Distance cannot be computed once you have more than three dimensions.
3
More dimensions always make similarity search more accurate.
4
It only means the computation gets slower.
check
ch 40 · unsupervised learning, embeddings, and recommenders
2/7
promptdojo
_
›
phase 08 · ai/ml engineering
›
ch 40 · unsupervised learning, embeddings, and recommenders
lesson 2 of 4 · dimensionality and distance traps
step 2 of 7 in this lesson
What is the "curse of dimensionality" for distance-based methods like nearest-neighbor?
1
As dimensions grow, points become nearly equidistant, so "nearest" stops being meaningful.
2
Distance cannot be computed once you have more than three dimensions.
3
More dimensions always make similarity search more accurate.
4
It only means the computation gets slower.
check
park a thought