Scaling Laws for Classical Machine Learning on Tabular Data: A Benchmark Study

• URL: http://arxiv.org/abs/2607.21866v2
• Date: Tue, 28 Jul 2026 17:34:47 GMT
• Title: Scaling Laws for Classical Machine Learning on Tabular Data: A Benchmark Study
• Abstract summary: 127 graduate students each ran a fixed protocol on 3 assigned datasets.<n>For 5 of 6 families, a single family-level exponent predicts each family's cross-dataset curves nearly as well as per-dataset exponents.<n>We release the aggregated curves, per-cell fits, and a practical data-requirement table for N* to reach target error 0.15.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Prior classical-ML learning-curve work fits power laws to tree, linear, and kernel models on tabular data, but at small scale: typically one curve, one team, a handful of cells. We present a distributed classroom-scale replication: 127 graduate students each ran a fixed protocol on 3 assigned datasets, drawn from 18 tabular classification and regression datasets and 6 model families (Boosting, Random Forest, SVM, Linear/Logistic, Ridge, Lasso), yielding 11,536 training runs and 1,648 fitted power-law curves of the form error(N) = a N^(-b) + c. Three findings. (1) Power laws fit: R^2 > 0.8 on 77.7% of cells, with tree ensembles dominating at full data (Boosting 50% of datasets, RandomForest 33%; linear models underperform on classification). (2) Approximate shared exponents within a model family: for 5 of 6 families, a single family-level exponent predicts each family's cross-dataset curves nearly as well as per-dataset exponents (R^2 gap < 0.011), though AIC favors the unconstrained fit and curve collapse is partial (32-58% of points within +/-0.5 dex). We frame this as approximate predictive compressibility, not dataset-independent universality; Lasso fails outright (negative control) and Ridge is fragile under leave-one-dataset-out. (3) Replicator-implementation variance: with random_state=42 fixed, independent re-implementations of the same protocol still differ by mean CV(b) = 0.144 on the fitted exponent -- not seed variance, but the spread induced by unconstrained parts of the protocol (preprocessing, encoding, missing-value handling). We release the aggregated curves, per-cell fits, and a practical data-requirement table for N* to reach target error 0.15.

Related papers

• Exact Degeneracy Under Balanced k-Shot Sampling:Consequences for Small-Sample Discriminant Analysis on LLM Embeddings [0.0]
  We show that balanced k-shot sampling induces an exact, provable degeneracy in a family of small-sample discriminant estimators.<n>We then evaluate the Kernelgression repaired framework on few-shot text classification on frozen embeddings.
  arXiv  Detail & Related papers  (2026-09-09T08:13:54Z)
• The Blending Ratio Is Not Where the Performance Is: Diagnosing Prototype Blending for Few-Shot Adaptation of Vision-Language Models [22.461983628330277]
  Many few-shot adaptation methods classify with a convex combination of the zero-shot text prototype and the mean of the K labelled image features.<n>We ask what is the right ratio, can it be estimated without validation data, and is finding it where the performance is?<n>Across 4,800 cells (ten datasets, five backbones including SigLIP, five shot counts, five seeds, four prompt tiers) this theoretically optimal ratio is a reliable estimate of the wrong quantity.<n> validation-free linear probes beat even the oracle-tuned blend: CLAP by +1.9 points and LP++ by +1.5 on average.
  arXiv  Detail & Related papers  (2026-08-23T15:02:33Z)
• TokenPrint: A Calibrated Token-Space Fingerprint for Language-Model Provenance [8.657685524975246]
  We introduce a training-free fingerprint based on the top-$k$ vocabulary projections of late hidden states elicited by 250 fixed knowledge probes.<n>We evaluate the method on 32 open-weight models from nine families (0.6B--32B) with documented relationships.
  arXiv  Detail & Related papers  (2026-08-08T13:56:39Z)
• Distribution-First Population Simulation: Collapse, Calibration, and Recall in Non-WEIRD LLM Persona Modeling [0.0]
  Synthetic-population tools run every individual as an independent large language model (LLM) agent.<n>We show that this paradigm has a basic failure mode, and we set a distribution-first corrective against it.
  arXiv  Detail & Related papers  (2026-07-17T15:45:57Z)
• Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables [2.1795865731681903]
  We study country-by-year time-series matrices, a common layout on open-data portals.<n>Their joint effect exceeds the sum of independent effects by up to +1.180 (2x2 factorial, 6 datasets).<n> bootstrap 95% CIs are strictly positive on 4/6 datasets, with strongest evidence on wide Type-II matrices.
  arXiv  Detail & Related papers  (2026-05-21T04:08:42Z)
• Amplifying, Not Learning: Fine-Tuned AI Text Detectors Amplify a Pretrained Direction [51.56484100374058]
  Text detectors amplify a pretrained typicality axis; they do not construct an AI-vs-human boundary.<n>On raw encoders before any task supervision, projecting onto centroid(AI)-centroid(HC3) achieves NYT-vs-HC3 AUROC 0.806/0.944/0.834 across three architectures.<n>On RoBERTa-base, raw projection exceeds fine-tuning; on RoBERTa-base, full fine-tuning reduces discrimination below raw on both fluent-formal populations tested.
  arXiv  Detail & Related papers  (2026-05-20T19:08:38Z)
• Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling [0.0]
  We measure the coupling between reasoning and truthfulness across 63 base models from 16 families.<n>We find a regime change invisible to loss curves below a family-dependent critical scale $N_c$, capabilities anticorrelate; above it, they cooperate.
  arXiv  Detail & Related papers  (2026-05-13T03:14:09Z)
• Sparse Regression under Correlation and Weak Signals: A Reproducible Benchmark of Classical and Bayesian Methods [1.6679662639178268]
  We benchmark six sparse regression methods on synthetic data.<n>Bayesian methods win on prediction error (MSE 72 vs. 108-267), and the Horseshoe delivers near-nominal 95% coverage (94.8%)<n>For variable selection, Lasso and Spike-and-Slab tie at F1 0.47, making Lasso the practical default when posteriors are not needed.
  arXiv  Detail & Related papers  (2026-04-04T15:46:44Z)
• Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models [81.62767292169225]
  We investigate knowledge forgetting in large language models with a focus on its generalisation.<n>We propose PerMU, a novel probability perturbation-based unlearning paradigm.<n>Experiments are conducted on a diverse range of datasets, including TOFU, Harry Potter, ZsRE, WMDP, and MUSE.
  arXiv  Detail & Related papers  (2025-02-27T11:03:33Z)
• Highly Adaptive Ridge [84.38107748875144]
  We propose a regression method that achieves a $n-2/3$ dimension-free L2 convergence rate in the class of right-continuous functions with square-integrable sectional derivatives. Har is exactly kernel ridge regression with a specific data-adaptive kernel based on a saturated zero-order tensor-product spline basis expansion. We demonstrate empirical performance better than state-of-the-art algorithms for small datasets in particular.
  arXiv  Detail & Related papers  (2024-10-03T17:06:06Z)
• Scaling Laws in Linear Regression: Compute, Parameters, and Data [86.48154162485712]
  We study the theory of scaling laws in an infinite dimensional linear regression setup.<n>We show that the reducible part of the test error is $Theta(-(a-1) + N-(a-1)/a)$.<n>Our theory is consistent with the empirical neural scaling laws and verified by numerical simulation.
  arXiv  Detail & Related papers  (2024-06-12T17:53:29Z)
• Pseudo-Labeling for Kernel Ridge Regression under Covariate Shift [1.3597551064547502]
  We learn a regression function with small mean squared error over a target distribution, based on unlabeled data from there and labeled data that may have a different feature distribution.<n>We propose to split the labeled data into two subsets, and conduct kernel ridge regression on them separately to obtain a collection of candidate models and an imputation model.<n>Our estimator achieves the minimax optimal error rate up to a polylogarithmic factor, and we find that using pseudo-labels for model selection does not significantly hinder performance.
  arXiv  Detail & Related papers  (2023-02-20T18:46:12Z)
• Few-Shot Non-Parametric Learning with Deep Latent Variable Model [50.746273235463754]
  We propose Non-Parametric learning by Compression with Latent Variables (NPC-LV) NPC-LV is a learning framework for any dataset with abundant unlabeled data but very few labeled ones. We show that NPC-LV outperforms supervised methods on all three datasets on image classification in low data regime.
  arXiv  Detail & Related papers  (2022-06-23T09:35:03Z)
• Test Set Sizing Via Random Matrix Theory [91.3755431537592]
  This paper uses techniques from Random Matrix Theory to find the ideal training-testing data split for a simple linear regression. It defines "ideal" as satisfying the integrity metric, i.e. the empirical model error is the actual measurement noise. This paper is the first to solve for the training and test size for any model in a way that is truly optimal.
  arXiv  Detail & Related papers  (2021-12-11T13:18:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.