AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Active Budget Allocation for Efficient Scaling Law Estimation via Surrogate-Guided Pruning

Viktoria Schram, Markus Hiller, Daniel Beck, Trevor Cohn

ICML 2026 · Architectures and efficiency › RNNs and normalization · approach: Neural scaling laws · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LMC surrogate-guided pruning to Successive HalvingSuccessive Halving · synthetic learning curves, 5 models, B=10^4 petaFLOPsmean relative improvement in minimum loss: +5.47 %Improved
use more candidate models (5 to 20) when adding LMC surrogate to SH (ablation)SH LMC · synthetic learning curves, B=10^4mean relative improvement over SH: 5.47 % -> 1.50 %Hurt
use Successive Halving instead of uniform allocation (ablation)budget allocation · synthetic learning curvesrelative loss: UA -10.17 % (5 models, B=10^2) vs SHImproved
use LMC surrogate instead of Deep Ensemble surrogates in SH (ablation)SH on nanoGPT · nanoGPT dataset, 20 models, B=10^4mean relative improvement over SH: 2.84 % vs 2.02 % (PL), 1.35 % (EXP), 1.80 % (MMF)Improved
add LMC surrogate to SH on real-world nanoGPT curvesSuccessive Halving · nanoGPT datasetmean relative improvement over SH: up to 2.84 %Improved
add LMC surrogate to SH, scaling law accuracySuccessive Halving · nanoGPT, AbC to Full Data SLarea between curves: 0.12 -> 0.11 (20 models, B=10^4); 0.09 -> 0.11 (5 models, B=10^4)No clear effect
add LMC surrogate to SH, loss regretSuccessive Halving · nanoGPT, 20 models, B=10^4loss regret: 0.67 -> 0.59Improved
add SH-based allocation versus exhaustive training of all modelsscaling law estimation · nanoGPT, 20 models, B=10^4compute cost saving: saving 98.70%Improved
add GP mean extrapolation after SH LMC allocation (ablation)SH LMC · Kaplan synthetic LCs, B=10^5AbC to ground truth SL: 2.17 -> 0.00 ± 0.00Improved
fit scaling law to GP UCB instead of GP mean (ablation)GP extrapolation · Kaplan synthetic LCs, B=10^5AbC to ground truth SL: 0.00 -> 0.53Hurt
add noise (white, Ornstein-Uhlenbeck, Brownian) to learning curves (ablation)SH LMC vs SH · synthetic learning curves, 5 and 20 modelsaverage minimum lossNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy