Capstractcscapstract:2611.00068

Computer Science > Machine Learning

Title:Thermal-State Thresholds for Early Stopping in Iterative Regression

Abstract:

We study whether the temperature of a cooling cup of tea can select a useful stopping time for gradient descent. A simulated cup follows Newtonian cooling, and training stops when its temperature crosses a calibrated threshold. For quadratic objectives, the normalized thermal trace exactly matches the optimization residual of a single Hessian eigendirection at a corresponding curvature. We also prove that, with fixed thermal conditions and update rate, the stopping rule is identical to a fixed computational budget. In synthetic linear regression, a threshold calibrated on 30 tasks attains mean excess prediction risk 0.416 on 100 new tasks from the same distribution, compared with 0.420 for per-task validation checkpoint selection and 0.964 at the final checkpoint. A matched clock reproduces the cup exactly. When the label-noise standard deviation doubles, cup risk increases to 1.181 versus 0.836 for validation. Altering the simulated cooling rate or update throughput changes the selected checkpoint. The results characterize thermal stopping as a transferable, externally timed regularizer and identify the distributional and timing assumptions on which it depends.

Comments:7 pages, 2 figures, 3 tables; synthetic regression and simulated cooling
Subjects:Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as:capstract:2611.00068 [cs.LG]
 (or capstract:2611.00068v1 [cs.LG] for this version)

Submission history

From: Elira Tavrennic
[v1] Fri, 20 Nov 2026 21:17:32 UTC (131 KB)

Current browse context:

stat
Change to browse by:

Export citation