The maximum of N distractor similarities
ProofAdvanced⏱ 30 minRAGmaximumsub-gaussianunion boundretrieval
Problem
Prove for independent uniform directions , and derive the high-probability version. Then compute the required query–document alignment for top-1 retrieval at and .
Write a complete argument. Every inequality you use should be one you could state precisely.
Solution
Opening this before you have a real attempt on paper is the fastest way to feel like you understand something you cannot reproduce.
Related problems
All problems| Solved | Problem | Field | Type | Difficulty | Time |
|---|---|---|---|---|---|
| Three similarities agree on the unit sphere | RAG | Proof | 15 min | ||
| When is the union bound tight? | Probability | Proof | 15 min | ||
| Generalisation for a finite class | Learning Theory | Proof | 30 min | ||
| A sufficient condition for correct retrieval | RAG | Proof | 30 min | ||
| Measure anisotropy in an embedding space run | RAG | Coding | 1 hour |