Pith. sign in

REVIEW 7 cited by

Accounting for Variance in Machine Learning Benchmarks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2103.03098 v1 pith:JL5RKDIC submitted 2021-03-01 cs.LG stat.ML

classification cs.LGstat.ML
keywords datavarianceestimatorinitializationlearningparameterresultssampling
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Strong empirical evidence that one machine-learning algorithm A outperforms another one B ideally calls for multiple trials optimizing the learning pipeline over sources of variation such as data sampling, data augmentation, parameter initialization, and hyperparameters choices. This is prohibitively expensive, and corners are cut to reach conclusions. We model the whole benchmarking process, revealing that variance due to data sampling, parameter initialization and hyperparameter choice impact markedly the results. We analyze the predominant comparison methods used today in the light of this variance. We show a counter-intuitive result that adding more sources of variation to an imperfect estimator approaches better the ideal estimator at a 51 times reduction in compute cost. Building on these results, we study the error rate of detecting improvements, on five different deep-learning tasks/architectures. This study leads us to propose recommendations for performance comparisons.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Where You Measure Decides What You Measure: Position Selection in Ablation-Based SAE Evaluation

    cs.LG 2026-08 accept novelty 7.0 of 10

    In ablation-based SAE evaluation, the measurement token is selected by the dictionary under test, and holding that token fixed collapses most of the variance that is usually attributed to differences between dictionaries.

  2. Governing Agentic AI in FinTech

    cs.CY 2026-08 conditional novelty 6.0 of 10

    Financial institutions can lose the ability to explain or reproduce agentic AI decisions even when the system is capable and seemingly stable; the paper names this a Verifiability Gap and dissects its mechanisms.

  3. Data Quality Profiling at Scale with Progressive Sampling: A Benchmark for Data-Centric AI Pipelines

    cs.DB 2026-07 conditional novelty 6.0 of 10

    Random uniform sampling dominates proxy-guided MCMC for data-quality profiling at scale, with up to 49x lower error on real datasets.

  4. Grokking Is Conditional and Fragile: A Fully-Tractable, Multi-Seed Study at 12K Parameters

    cs.LG 2026-07 accept novelty 6.0 of 10

    In a fully tractable 12K Llama-style model, grokking is a conditional fragile phase transition gated by coverage (tracking modulus more than structure), weight decay, and floating-point reduction order, so evidence mu...

  5. Beginning with You: Perceptual-Initialization Improves Vision-Language Representation and Alignment

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Perceptually initializing a CLIP vision encoder with NIGHTS triplet judgments before YFCC15M contrastive training improves zero-shot accuracy and retrieval over an identical random-start baseline.

  6. Evaluation of Seismic Artificial Intelligence with Uncertainty

    cs.LG 2025-01 conditional novelty 6.0 of 10

    A seismic AI evaluation framework jointly estimates training and data uncertainty and learning efficiency, and shows no single PhaseNet training approach wins at all data budgets.

  7. SketchMamba: A Lightweight State-Space Model for Joint Progressive Sketch Classification and Stroke Auto-Completion

    cs.CV 2026-07 conditional novelty 5.0 of 10

    Dense per-step class loss on a 1.55M-parameter causal Mamba unifies progressive sketch recognition (94.93% final Top-1, 0.706 progressive AUC) and stroke auto-completion on 58-class Quick, Draw!.

Pith tools