Unverified50% confidenceFactExact time
The paper proves that a continuous plug-in estimator based on kernel density estimation achieves the minimax rate n^{-s/(d+2s)} when the support manifolds are known.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThrough a coupled tensor gauge and physically localized packets, the researchers determined a sharp limiting nonlinear inverse coefficient, and proved that the optimal "density-only" gain is exactly the reciprocal of this coefficient.62% similarUnverifiedThe paper provides an estimator at the algorithmic level that uses pooled isotonic fitting plus a per-problem correction term, and this estimator does not need to know τ, V or μ in advance, achieving the minimax rate as long as n≥2 and m≥3.62% similarUnverifiedKeskar等人2017年的研究表明，平坦极小值对应更好的泛化性能60% similarUnverifiednDCG@k is calculated as discounted cumulative gain (DCG) divided by ideal DCG (ideal DCG), with a value range from 0 to 1, and supports graded relevance.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023426API
curl https://kongchang.com/api/v1/knowledge/claims/1023426MCP
get_claim(id=1023426)