A Selection Premium Decomposition for the Expected Maximum of Random Walks
When $K$ models are evaluated on the same validation set of size $n$, the selected winner's apparent performance is biased upward. Suppose $K$ models are evaluated on a shared sequence of i.i.d. observations $X_1,\dots, X_n$, where model $k$ achieves response $f_k(X_i)$ with mean $μ_k = \mathbb E[f_k(X)]$. Writing $Y_{i,k} = f_k(X_i)-μ_k$ for the centered increment and $S_{n,k} = \sum_{i=1}^n Y_{i,k}$ for the centered cumulative score, the expected maximum satisfies $0\le\mathbb E\bigl[\max_k S_{n,k}\bigr] = \sum_{i=1}^n \mathbb E\bigl[\varphi_K(S_{i-1})\bigr]$ where $\varphi_K(u) =…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2602.19481
curl -X POST /api/v1/records/arxiv/2602.19481/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2602.19481/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'