Abstract:We study long-horizon deployment of a frozen predictor under dynamic covariate shift. A time-domain Poincare inequality first reduces temporal risk volatility to derivative energy. A Jacobian-velocity theorem then supplies the corresponding pathwise control. Given explicit regularity and domination assumptions, the theorem identifies directional tangent energy along the deployment path as the governing quantity. Under low-rank drift, that quantity reduces to directional Jacobian energy in the drift subspace, motivating drift-aligned tangent regularization (DTR) and a matched monitoring proxy. Rather than smoothing the network isotropically, DTR penalizes sensitivity only along estimated drift directions. We validate the theorem-to-method pipeline in four experiments: a synthetic benchmark for the time-domain inequality, a controlled synthetic comparison against isotropic Jacobian regularization, and two frozen-deployment studies on the UCI Air Quality and Tetouan power-consumption datasets. DTR reduces risk volatility and directional gain in the controlled low-rank regime and beats isotropic smoothing there. It also gives validation-selected deployment gains on both real datasets, with the Air Quality subspace estimated from target-orthogonal sensor motion. Moderate drift-subspace misspecification is tolerable while orthogonal misspecification largely removes the benefit.
| Comments: | 8 pages, 4 figures, 4 tables |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG) |
| MSC classes: | 68T05, 68T07, 62R07, 62M10 |
| ACM classes: | I.2.6; G.3 |
| Cite as: | arXiv:2605.04932 [stat.ML] |
| (or arXiv:2605.04932v2 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2605.04932 arXiv-issued DOI via DataCite |
From: Jonathan Landers [view email]
[v1]
Wed, 6 May 2026 13:57:03 UTC (82 KB)
[v2]
Tue, 26 May 2026 16:04:20 UTC (82 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。