An empirical comparison of four deep learning fault localisation tools on 22 faults shows low accuracy under a single ground truth, with measured recall rising when alternative equivalent fixes are included.
DeepDiagnosis: automatically diagnosing faults and recommending actionable fixes in deep learning programs,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SE 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
An Empirical Study of Fault Localisation Techniques for Deep Learning
An empirical comparison of four deep learning fault localisation tools on 22 faults shows low accuracy under a single ground truth, with measured recall rising when alternative equivalent fixes are included.