Unverified50% confidenceFactExact time
SASORI achieves an optimal trade-off between region estimation error and cumulative regret.
1
Sources
50%
Confidence
Long-term
Relevance
11/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedInference regret measures the gap between the global optimum value and the function value of the point the algorithm finally returns.68% similarUnverified数值实验表明三项误差中score误差项占据主导,整体TV误差与score误差呈线性关系65% similarUnverifiedThe sample complexity of SASORI matches the lower bound up to logarithmic factors and is therefore minimax optimal.65% similarUnverifiedThe gap between the oracle upper bound (assuming selected errors can be perfectly corrected) and the actual gain essentially reflects the mismatch between error fixability and error detectability.65% similarUnverifiedTotal prediction error can be decomposed into bias² + variance + irreducible error.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1158425API
curl https://kongchang.com/api/v1/knowledge/claims/1158425MCP
get_claim(id=1158425)