Mutation Testing of Deep Learning Model


Dr. Muhammad Iqbal Hossain (MIH)

Associate Professor

iqbal.hossain@bracu.ac.bd

Synopsis

 

Deep neural networks (DNNs) are increasingly finding use in a variety of real-world applications, including speech recognition, image processing, and natural language processing. However, in terms of test data quality and model robustness, there is currently a lack of tool support for DNN testing. 


Relevance of the Topic

 

Currently most of the works are focused on DL models, how to increase accuracy etc. Now it is the time to test the quality of the DL model.

 


Future Research/Scope

 

Student will be learn the internals of DL models and mutation testing. It can be used for testing DL applications.

 


Skills Learned

 

Deep learning algorithm, Python, testing

 


Relevant courses to the topic

 

Neural Networks, Software Engineering

 


Reading List

 

1. DeepMutation: Mutation Testing of Deep Learning Systems
2. DeepMutation++: A Mutation Testing Framework for Deep Learning Systems

 



©2024 BracU CSE Department