MétaCan
Menu
Back to cohort
Record W4287080893 · doi:10.48550/arxiv.2107.04863

HOMRS: High Order Metamorphic Relations Selector for Deep Neural\n Networks

2021· preprint· en· W4287080893 on OpenAlex
Florian Tambon, Giulio Antoniol, Foutse Khomh

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

VenuearXiv (Cornell University) · 2021
Typepreprint
Languageen
FieldComputer Science
TopicAdversarial Robustness in Machine Learning
Canadian institutionsPolytechnique Montréal
Fundersnot available
KeywordsComputer scienceGeneralizationMNIST databaseSet (abstract data type)Artificial intelligenceArtificial neural networkScheme (mathematics)Machine learningOrder (exchange)Code (set theory)Path (computing)ExploitDeep learningTheoretical computer scienceProgramming languageMathematics

Abstract

fetched live from OpenAlex

Deep Neural Networks (DNN) applications are increasingly becoming a part of\nour everyday life, from medical applications to autonomous cars. Traditional\nvalidation of DNN relies on accuracy measures, however, the existence of\nadversarial examples has highlighted the limitations of these accuracy\nmeasures, raising concerns especially when DNN are integrated into\nsafety-critical systems.\n In this paper, we present HOMRS, an approach to boost metamorphic testing by\nautomatically building a small optimized set of high order metamorphic\nrelations from an initial set of elementary metamorphic relations. HOMRS'\nbackbone is a multi-objective search; it exploits ideas drawn from traditional\nsystems testing such as code coverage, test case, path diversity as well as\ninput validation.\n We applied HOMRS to MNIST/LeNet and SVHN/VGG and we report evidence that it\nbuilds a small but effective set of high-order transformations that generalize\nwell to the input data distribution. Moreover, comparing to similar generation\ntechnique such as DeepXplore, we show that our distribution-based approach is\nmore effective, generating valid transformations from an uncertainty\nquantification point of view, while requiring less computation time by\nleveraging the generalization ability of the approach.\n

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)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.915
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.001
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.002
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0020.002
Research integrity0.0010.001
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 head0.044
GPT teacher head0.192
Teacher spread0.148 · 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