MétaCan
Menu
Back to cohort
Record W4226000139 · doi:10.1093/imaiai/iaae002

PACMAN: PAC-style bounds accounting for the Mismatch between Accuracy and Negative log-loss

2024· preprint· en· W4226000139 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueInformation and Inference A Journal of the IMA · 2024
Typepreprint
Languageen
FieldComputer Science
TopicMachine Learning and Data Classification
Canadian institutionsÉcole de Technologie SupérieureMcGill University
FundersUniversidad de Buenos AiresConsejo Nacional de Investigaciones Científicas y Técnicas
KeywordsGeneralizationComputer scienceAlgorithmFunction (biology)Metric (unit)Generalization errorCross entropyMathematicsArtificial intelligenceArtificial neural networkPattern recognition (psychology)

Abstract

fetched live from OpenAlex

Abstract The ultimate performance of machine learning algorithms for classification tasks is usually measured in terms of the empirical error probability (or accuracy) using a testing dataset, although these algorithms are optimized through the minimization of a typically different—more convenient—loss function using a training set. For classification tasks, this loss function is often the negative log-loss that yields the well-known cross-entropy risk that is typically better behaved (in terms of numerical behavior) than the 0-1 loss. Conventional studies on the generalization error do not usually take into account the underlying mismatch between losses at training and testing phases. In this work, we introduce a theoretical analysis based on a pointwise probably approximately correct (PAC) approach over the generalization gap considering the mismatch of testing on the accuracy metric and training on the negative log-loss, referred to as PACMAN. Building on the fact that the resulting mismatch can be written as a likelihood ratio, concentration inequalities can be used to obtain insights into the generalization gap in terms of PAC bounds, which depend on some meaningful information-theoretic quantities. An analysis of the obtained bounds and a comparison with available results in the literature is also provided.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesScholarly communication
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Other design · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.947
Threshold uncertainty score0.999

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0020.002
Open science0.0010.001
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.027
GPT teacher head0.312
Teacher spread0.285 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it