Mehil B. Shah

PhD Student at Dalhousie University

prof_pic.jpg

Faculty of CS

Dalhousie University

Halifax, NS

I am a PhD Student in the Faculty of Computer Science at Dalhousie University. I am currently being advised by Prof. Masud Rahman, and Prof. Foutse Khomh. My primary research interests are at the interesection of Software Engineering and Deep Learning, with a focus on secure, reliable and explainable deep learning systems. My current research focuses on understanding the behaviour of bugs in deep learning systems, and trying to improve reproducibility of deep learning bugs.

In the recent times, I have been particularly intrigued by the role of data quality and data bugs in deep learning systems. These bugs originating from training data issues can significantly influence the performance and functionality of deep learning models. Given the widespread occurrence and critical impact of data bugs, there is a clear need for effective techniques to debug and address these issues within the training data.

news

May, 2024 We received Major Revision for our paper “Towards Enhancing the Reproducibility of Deep Learning Bugs: An Empirical Study” from EMSE.
Apr, 2024 I have successfully completed the coursework for my PhD degree. I will move on to the Research Aptitude Defence later this year.
Apr, 2024 I will be serving as the Teaching Assistant and/or Marker for the courses CSCI x691, and CSCI 6409 in the Summer 2024.
Jan, 2024 I submitted two papers to EMSE on Deep Learning Bug Reproducibility and Deep Learning Bug Localization, one as first author and the other as second author. Preprints for both papers will be available soon.
Dec, 2023 I will serve as a teaching assistant for two courses in the Winter 2024 term: CSCI x691 and CSCI 5308.
Dec, 2023 I served on the Junior Program Committee for the 2024 Mining Software Repositories (MSR) conference, where I reviewed 5 papers on a wide range of topics.
Nov, 2023 Served as a sub-reviewer for the IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) 2024.