Unverified50% confidenceBenchmarkExact time
Under the total variation metric, the average error is Õ_p(n^(−β/(d_x+d_y+2β))).
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUnder the assumption that the true mapping is uniformly random, the paper proves a lower bound on the error rate of E ≥ (M/N)δ*(B/M) + (1-M/N)(1-1/K).79% similarUnverifiedThe error introduced by uniform quantization approximately follows a uniform distribution, with a variance of Δ²/12 (where Δ is the quantization step size).77% similarUnverifiedOn a well-conditioned zero-predictor family, the optimal B-bit compression yields a mean single-step update error on the order of d^{-1}·2^{-2B/s_d}, where s_d=d(d+1)/2.75% similarUnverifiedIn i.i.d. sampling, the statistical error for n samples typically decays at the rate of 1/√n.74% similarUnverifiedThe paper proves that the learned generative distribution satisfies a finite-sample error bound in Wasserstein-p distance: W_p ≲ n^(-1/d) + n^(-1/(2p))(log(1/ξ))^(1/(2p)).72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002242API
curl https://kongchang.com/api/v1/knowledge/claims/1002242MCP
get_claim(id=1002242)