Taxonomy of Real Faults in Deep Learning Systems

10/24/2019
by   Gunel Jahangirova, et al.
0

The growing application of deep neural networks in safety-critical domains makes the analysis of faults that occur in such systems of enormous importance. In this paper we introduce a large taxonomy of faults in deep learning (DL) systems. We have manually analysed 1059 artefacts gathered from GitHub commits and issues of projects that use the most popular DL frameworks (TensorFlow, Keras and PyTorch) and from related Stack Overflow posts. Structured interviews with 20 researchers and practitioners describing the problems they have encountered in their experience have enriched our taxonomy with a variety of additional faults that did not emerge from the other two sources. Our final taxonomy was validated with a survey involving an additional set of 21 developers, confirming that almost all fault categories (13/15) were experienced by at least 50

READ FULL TEXT
research
01/01/2021

Faults in Deep Reinforcement Learning Programs: A Taxonomy and A Detection Approach

A growing demand is witnessed in both industry and academia for employin...
research
12/13/2020

Fault Injectors for TensorFlow: Evaluation of the Impact of Random Hardware Faults on Deep CNNs

Today, Deep Learning (DL) enhances almost every industrial sector, inclu...
research
07/28/2021

Clones in Deep Learning Code: What, Where, and Why?

Deep Learning applications are becoming increasingly popular. Developers...
research
06/05/2021

An Empirical Study on Tensor Shape Faults in Deep Learning Systems

Software developers frequently adopt deep learning (DL) libraries to inc...
research
09/14/2019

FfDL : A Flexible Multi-tenant Deep Learning Platform

Deep learning (DL) is becoming increasingly popular in several applicati...
research
07/26/2023

Sources of Opacity in Computer Systems: Towards a Comprehensive Taxonomy

Modern computer systems are ubiquitous in contemporary life yet many of ...
research
09/13/2023

Data Pipeline Quality: Influencing Factors, Root Causes of Data-related Issues, and Processing Problem Areas for Developers

Data pipelines are an integral part of various modern data-driven system...

Please sign up or login with your details

Forgot password? Click here to reset