From Spectra to Joint Schedules in LLM Pre-training: 3+3(+2) Scaling-Law Regimes

arXiv:2609.40148v1 Announce Type: new Abstract: Power-law learning curves are often treated as fixed properties of a model and its data, although learning-rate and batch-size schedules can change the observed loss. We study this dependence in noisy online SGD with linear random features. Conditional on the representation, an exact Volterra equation separates two response components: a forcing term that propagates unresolved target error and a memory kernel that propagates stochastic-error injections. We prove that either component follows a power law if and only if its cumulative weighted spectral mass has the corresponding low-spectrum scaling; individual eigenvalues and target coefficients need not obey coordinatewise power laws. Under a joint schedule, intrinsic time T_t=sum_{s<t}eta_s controls optimization progress, while r_t=B_t/eta_t controls noise injection. Their interaction yields sharp conditions under which a schedule preserves, changes, or destroys the clean power law, together with a memory ceiling on noise reduction. The power-law random-feature model realizes this mechanism in 3+3(+2) propagation regimes with phase-dependent compute rates. Controlled nanoGPT experiments show that (1) learning-rate and batch-size schedules with matched B/eta paths are nearly equivalent in intrinsic time, (2) a forcing-memory surrogate accurately predicts loss across schedules, and (3) its fitted exponents across real-world datasets identify the regime of LLMs in 3+3(+2) map.

Related

• Efficiently Estimating Optimal Hyperparameter Scaling Laws through Power-Law Entropy Search
• A Nationwide Japanese Medical Claims Foundation Model: Balancing Model Scaling and Task-Specific Computational Efficiency
• In-Context Learning Under Regime Change

Source: arXiv cs.LG | 2026-10-01