MétaCan
Menu
Back to cohort
Record W3196967454 · doi:10.1109/qrs54544.2021.00116

The Challenge of Reproducible ML: An Empirical Study on The Impact of Bugs

2021· article· en· W3196967454 on OpenAlexaff
Emilio Rivera-Landos, Foutse Khomh, Amin Nikanjam

Bibliographic record

Venue2021 IEEE 21st International Conference on Software Quality, Reliability and Security (QRS) · 2021
Typearticle
Languageen
FieldComputer Science
TopicMachine Learning and Data Classification
Canadian institutionsPolytechnique Montréal
Fundersnot available
KeywordsComputer scienceDeterminismInferenceSoftware bugReproducibilitySoftwareMachine learningArtificial intelligenceFace (sociological concept)Causal inferenceSoftware engineeringProgramming languageStatistics

Abstract

fetched live from OpenAlex

Reproducibility is a crucial requirement in scientific research. When results of research studies and scientific papers have been found difficult or impossible to reproduce, we face a challenge which is called reproducibility crisis. Although the demand for reproducibility in Machine Learning (ML) is acknowledged in the literature, a main barrier is inherent non-determinism in ML training and inference. In this paper, we establish the fundamental factors that cause non-determinism in ML systems. A framework, ReproduceML, is then introduced for deterministic evaluation of ML experiments in a real, con-trolled environment. ReproduceML allows researchers to investigate software configuration effects on ML training and inference. Using ReproduceML, we run a case study: investigation of the impact of bugs inside ML libraries on performance of ML experiments. This study attempts to quantify the impact that the occurrence of bugs in a popular ML framework, PyTorch, has on the performance of trained models. To do so, a comprehensive methodology is proposed to collect buggy versions of ML libraries and run deterministic ML experiments using ReproduceML. Our initial finding is that there is no evidence based on our limited dataset to show that bugs which occurred in PyTorch do affect the performance of trained models. The proposed methodology as well as ReproduceML can be employed for further research on non-determinism and bugs.

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.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.067
metaresearch head score (Gemma)0.410
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Reproducibility · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.933
Threshold uncertainty score0.353

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0670.410
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.003
Science and technology studies0.0020.006
Scholarly communication0.0040.007
Open science0.0030.004
Research integrity0.0020.004
Insufficient payload (model declined to judge)0.0020.001

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 head0.114
GPT teacher head0.420
Teacher spread0.306 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designObservational
DomainReproducibility
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations5
Published2021
Admission routes1
Has abstractyes

Explore more

Same venue2021 IEEE 21st International Conference on Software Quality, Reliability and Security (QRS)Same topicMachine Learning and Data ClassificationFrench-language works237,207