MétaCan
Menu
Back to cohort
Record W4280562623 · doi:10.1145/3529318

Testing Feedforward Neural Networks Training Programs

2022· article· en· W4280562623 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.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueACM Transactions on Software Engineering and Methodology · 2022
Typearticle
Languageen
FieldComputer Science
TopicAdversarial Robustness in Machine Learning
Canadian institutionsPolytechnique Montréal
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsComputer scienceDebuggingHyperparameterArtificial neural networkMachine learningArtificial intelligenceSoftwareDeep neural networksTraining (meteorology)Deep learningTest dataSoftware engineeringProgramming language

Abstract

fetched live from OpenAlex

At present, we are witnessing an increasing effort to improve the performance and trustworthiness of Deep Neural Networks (DNNs), with the aim to enable their adoption in safety critical systems such as self-driving cars or aircraft collision-avoidance systems. Multiple testing techniques are proposed to generate test cases that can expose inconsistencies in the behavior of DNN models. These techniques assume implicitly that the training program is bug-free and appropriately configured. However, satisfying this assumption for a novel problem requires significant engineering work to prepare the data, design the DNN, implement the training program, and tune the hyperparameters to produce the model for which current automated test data generators search for corner-case behaviors. All these model training steps can be error prone. Therefore, it is crucial to detect and correct errors throughout all the engineering steps of DNN-based software systems and not only on the resulting DNN model. In this article, we gather a catalog of training issues and based on their symptoms and their effects on the behavior of the training program, we propose practical verification routines to detect the aforementioned issues, automatically, by continuously validating that some important properties of the learning dynamics hold during the training. Then, we design TheDeepChecker , an end-to-end property-based debugging approach for DNN training programs and implement it as a TensorFlow-based library. As an empirical evaluation, we conduct a case study to assess the effectiveness of TheDeepChecker on synthetic and real-world buggy DL programs and compare its performance to that of the Amazon SageMaker Debugger ( SMD ). Results show that TheDeepChecker ’s on-execution validation of DNN-based program’s properties through three sequential phases (pre-, on-, and post-fitting) succeeds in revealing several coding bugs and system misconfigurations errors early on and at a low cost. Moreover, our property-based approach outperforms the SMD ’s offline rules verification on training logs in terms of detection accuracy for unstable learning issues and coverage of additional DL 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.

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 categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.390
Threshold uncertainty score0.869

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0010.000
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 head0.120
GPT teacher head0.306
Teacher spread0.186 · 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