Unverified50% confidenceFactExact time
The paper theoretically proves that AMTP's prediction error translates into throughput regret relative to the best horizon for each context, and that this regret is sublinear when online gradient descent is used.
1
Sources
50%
Confidence
Long-term
Relevance
10/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTotal prediction error can be decomposed into bias² + variance + irreducible error.69% similarUnverifiedAlgorithms such as online gradient descent can guarantee O(√T) regret under convex or approximately convex assumptions.69% similarVerifiedIn autoregressive prediction, errors at each step accumulate into the next, causing long-range predictions to degrade rapidly.69% similarUnverifiedPrior error refers to the gap between the synthetic prior and the true data distribution, context error refers to insufficient information caused by finite context samples, and amortization error refers to the approximation bias of the network itself in approximating the target PPD.68% similarVerifiedGradient descent iteratively adjusts weights in the opposite direction of the loss function's gradient, causing the prediction error to converge progressively.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1078992API
curl https://kongchang.com/api/v1/knowledge/claims/1078992MCP
get_claim(id=1078992)