Portrait of Foutse Khomh

Foutse Khomh

Associate Academic Member
Canada CIFAR AI Chair
Professor, Polytechnique Montréal, Department of Computer Engineering and Software Engineering
Research Topics
Data Mining
Deep Learning
Distributed Systems
Generative Models
Learning to Program
Natural Language Processing
Reinforcement Learning
Website
Google Scholar
LinkedIn
X

Biography

Foutse Khomh is a full professor of software engineering at Polytechnique Montréal, a Canada CIFAR AI Chair – Trustworthy Machine Learning Software Systems, and an FRQ-IVADO Research Chair in Software Quality Assurance for Machine Learning Applications. Khomh completed a PhD in software engineering at Université de Montréal in 2011, for which he received an Award of Excellence. He was also awarded a CS-Can/Info-Can Outstanding Young Computer Science Researcher Prize in 2019.

His research interests include software maintenance and evolution, machine learning systems engineering, cloud engineering, and dependable and trustworthy ML/AI. His work has received four Ten-year Most Influential Paper (MIP) awards, and six Best/Distinguished Paper Awards. He has served on the steering committee of numerous organizations in software engineering, including SANER (chair), MSR, PROMISE, ICPC (chair), and ICSME (vice-chair). He initiated and co-organized Polytechnique Montréal‘s Software Engineering for Machine Learning Applications (SEMLA) symposium and the RELENG (release engineering) workshop series.

Khomh co-founded the NSERC CREATE SE4AI: A Training Program on the Development, Deployment and Servicing of Artificial Intelligence-based Software Systems, and is a principal investigator for the DEpendable Explainable Learning (DEEL) project.

He also co-founded Confiance IA, a Quebec consortium focused on building trustworthy AI, and is on the editorial board of multiple international software engineering journals, including IEEE Software, EMSE and JSEP. He is a senior member of IEEE.

Current Students

Nanda Assobjio Brice Yvan
Master's Research - Polytechnique Montréal
Google Scholar
Gabriel Laberge
PhD - Polytechnique Montréal
Github
Google Scholar
forough majidi
PhD - Polytechnique Montréal
Website
Google Scholar
Mo Malekpour
Master's Research - Polytechnique Montréal
mohammadhossein.malekpour@gmail.com
Website
Github
Google Scholar
Mohamed Amine Merzouk
Postdoctorate - Polytechnique Montréal
Co-supervisor :
Sarath Chandar
Website
Github
Google Scholar
Arghavan Moradi Dakhel
Postdoctorate - Polytechnique Montréal
Website
Github
Google Scholar
Khouloud Oueslati
Master's Research - Polytechnique Montréal
Elnathan Tiokou Tiokou Fangang
PhD - Polytechnique Montréal
Github
Ben Braiek Yasmine
Master's Research - Polytechnique Montréal

Publications

Deep Learning Model Reuse in the HuggingFace Community: Challenges, Benefit and Trends
Mina Taraghi
Gianolli Dorcelus
Armstrong Foundjem
Florian Tambon
Foutse Khomh
The ubiquity of large-scale Pre-Trained Models (PTMs) is on the rise, sparking interest in model hubs, and dedicated platforms for hosting P… (see more)TMs. Despite this trend, a comprehensive exploration of the challenges that users encounter and how the community leverages PTMs remains lacking. To address this gap, we conducted an extensive mixed-methods empirical study by focusing on discussion forums and the model hub of HuggingFace, the largest public model hub. Based on our qualitative analysis, we present a taxonomy of the challenges and benefits associated with PTM reuse within this community. We then conduct a quantitative study to track model-type trends and model documentation evolution over time. Our findings highlight prevalent challenges such as limited guidance for beginner users, struggles with model output comprehensibility in training or inference, and a lack of model understanding. We also identified interesting trends among models where some models maintain high upload rates despite a decline in topics related to them. Additionally, we found that despite the introduction of model documentation tools, its quantity has not increased over time, leading to difficulties in model comprehension and selection among users. Our study sheds light on new challenges in reusing PTMs that were not reported before and we provide recommendations for various stakeholders involved in PTM reuse.
2024-03-12
2024 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) (published)
doi.org
arxiv.org
Refining GPT-3 Embeddings with a Siamese Structure for Technical Post Duplicate Detection
Xingfang Wu
Heng Li
Nobukazu Yoshioka
Hironori Washizaki
Foutse Khomh
2024-03-12
2024 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) (published)
doi.org
arxiv.org
Trained Without My Consent: Detecting Code Inclusion In Language Models Trained on Code
Vahid Majdinasab
Amin Nikanjam
Foutse Khomh
Code auditing ensures that the developed code adheres to standards, regulations, and copyright protection by verifying that it does not cont… (see more)ain code from protected sources. The recent advent of Large Language Models (LLMs) as coding assistants in the software development process poses new challenges for code auditing. The dataset for training these models is mainly collected from publicly available sources. This raises the issue of intellectual property infringement as developers’ codes are already included in the dataset. Therefore, auditing code developed using LLMs is challenging, as it is difficult to reliably assert if an LLM used during development has been trained on specific copyrighted codes, given that we do not have access to the training datasets of these models. Given the non-disclosure of the training datasets, traditional approaches such as code clone detection are insufficient for asserting copyright infringement. To address this challenge, we propose a new approach, TraWiC; a model-agnostic and interpretable method based on membership inference for detecting code inclusion in an LLM’s training dataset. We extract syntactic and semantic identifiers unique to each program to train a classifier for detecting code inclusion. In our experiments, we observe that TraWiC is capable of detecting 83.87% of codes that were used to train an LLM. In comparison, the prevalent clone detection tool NiCad is only capable of detecting 47.64%. In addition to its remarkable performance, TraWiC has low resource overhead in contrast to pair-wise clone detection that is conducted during the auditing process of tools like CodeWhisperer reference tracker, across thousands of code snippets.
2024-02-14
ArXiv (preprint)
doi.org
arxiv.org
Trained Without My Consent: Detecting Code Inclusion In Language Models Trained on Code
Vahid Majdinasab
Amin Nikanjam
Foutse Khomh
Code auditing ensures that the developed code adheres to standards, regulations, and copyright protection by verifying that it does not cont… (see more)ain code from protected sources. The recent advent of Large Language Models (LLMs) as coding assistants in the software development process poses new challenges for code auditing. The dataset for training these models is mainly collected from publicly available sources. This raises the issue of intellectual property infringement as developers’ codes are already included in the dataset. Therefore, auditing code developed using LLMs is challenging, as it is difficult to reliably assert if an LLM used during development has been trained on specific copyrighted codes, given that we do not have access to the training datasets of these models. Given the non-disclosure of the training datasets, traditional approaches such as code clone detection are insufficient for asserting copyright infringement. To address this challenge, we propose a new approach, TraWiC; a model-agnostic and interpretable method based on membership inference for detecting code inclusion in an LLM’s training dataset. We extract syntactic and semantic identifiers unique to each program to train a classifier for detecting code inclusion. In our experiments, we observe that TraWiC is capable of detecting 83.87% of codes that were used to train an LLM. In comparison, the prevalent clone detection tool NiCad is only capable of detecting 47.64%. In addition to its remarkable performance, TraWiC has low resource overhead in contrast to pair-wise clone detection that is conducted during the auditing process of tools like CodeWhisperer reference tracker, across thousands of code snippets.
2024-02-14
ArXiv (preprint)
doi.org
arxiv.org
ChatGPT vs LLaMA: Impact, Reliability, and Challenges in Stack Overflow Discussions
Leuson Da Silva
Jordan Samhi
Foutse Khomh
Since its release in November 2022, ChatGPT has shaken up Stack Overflow, the premier platform for developers' queries on programming and so… (see more)ftware development. Demonstrating an ability to generate instant, human-like responses to technical questions, ChatGPT has ignited debates within the developer community about the evolving role of human-driven platforms in the age of generative AI. Two months after ChatGPT's release, Meta released its answer with its own Large Language Model (LLM) called LLaMA: the race was on. We conducted an empirical study analyzing questions from Stack Overflow and using these LLMs to address them. This way, we aim to (ii) measure user engagement evolution with Stack Overflow over time; (ii) quantify the reliability of LLMs' answers and their potential to replace Stack Overflow in the long term; (iii) identify and understand why LLMs fails; and (iv) compare LLMs together. Our empirical results are unequivocal: ChatGPT and LLaMA challenge human expertise, yet do not outperform it for some domains, while a significant decline in user posting activity has been observed. Furthermore, we also discuss the impact of our findings regarding the usage and development of new LLMs.
2024-02-13
ArXiv (preprint)
doi.org
arxiv.org
Deep Learning Model Reuse in the HuggingFace Community: Challenges, Benefit and Trends
Mina Taraghi
Gianolli Dorcelus
Armstrong Foundjem
Florian Tambon
Foutse Khomh
The ubiquity of large-scale Pre-Trained Models (PTMs) is on the rise, sparking interest in model hubs, and dedicated platforms for hosting P… (see more)TMs. Despite this trend, a comprehensive exploration of the challenges that users encounter and how the community leverages PTMs remains lacking. To address this gap, we conducted an extensive mixed-methods empirical study by focusing on discussion forums and the model hub of HuggingFace, the largest public model hub. Based on our qualitative analysis, we present a taxonomy of the challenges and benefits associated with PTM reuse within this community. We then conduct a quantitative study to track model-type trends and model documentation evolution over time. Our findings highlight prevalent challenges such as limited guidance for beginner users, struggles with model output comprehensibility in training or inference, and a lack of model understanding. We also identified interesting trends among models where some models maintain high upload rates despite a decline in topics related to them. Additionally, we found that despite the introduction of model documentation tools, its quantity has not increased over time, leading to difficulties in model comprehension and selection among users. Our study sheds light on new challenges in reusing PTMs that were not reported before and we provide recommendations for various stakeholders involved in PTM reuse.
2024-01-24
ArXiv (preprint)
doi.org
arxiv.org
Towards Enhancing the Reproducibility of Deep Learning Bugs: An Empirical Study
Mehil B. Shah
Mohammad Masudur Rahman
Foutse Khomh
Context: Deep learning has achieved remarkable progress in various domains. However, like any software system, deep learning systems contain… (see more) bugs, some of which can have severe impacts, as evidenced by crashes involving autonomous vehicles. Despite substantial advancements in deep learning techniques, little research has focused on reproducing deep learning bugs, which is an essential step for their resolution. Existing literature suggests that only 3% of deep learning bugs are reproducible, underscoring the need for further research. Objective: This paper examines the reproducibility of deep learning bugs. We identify edit actions and useful information that could improve the reproducibility of deep learning bugs. Method: First, we construct a dataset of 668 deep-learning bugs from Stack Overflow and GitHub across three frameworks and 22 architectures. Second, out of the 668 bugs, we select 165 bugs using stratified sampling and attempt to determine their reproducibility. While reproducing these bugs, we identify edit actions and useful information for their reproduction. Third, we used the Apriori algorithm to identify useful information and edit actions required to reproduce specific types of bugs. Finally, we conducted a user study involving 22 developers to assess the effectiveness of our findings in real-life settings. Results: We successfully reproduced 148 out of 165 bugs attempted. We identified ten edit actions and five useful types of component information that can help us reproduce the deep learning bugs. With the help of our findings, the developers were able to reproduce 22.92% more bugs and reduce their reproduction time by 24.35%. Conclusions: Our research addresses the critical issue of deep learning bug reproducibility. Practitioners and researchers can leverage our findings to improve deep learning bug reproducibility.
2024-01-05
ArXiv (preprint)
doi.org
arxiv.org
AITA: AI trustworthiness assessment
Bertrand Braunschweig
Stefan Buijsman
Faicel Chamroukhi
Fredrik Heintz
Foutse Khomh
Juliette Mattioli
Maximilian Poretschkin
2024-01-03
AI and Ethics (published)
doi.org
AmbieGen at the SBFT 2024 Tool Competition - CPS-UAV Track
Dmytro Humeniuk
Foutse Khomh
2024-01-01
SBFT@ICSE (published)
doi.org
AmbieGenVAE at the SBFT 2024 Tool Competition - Cyber-Physical Systems Track
Dmytro Humeniuk
Foutse Khomh
2024-01-01
SBFT@ICSE (published)
doi.org
Common Challenges of Deep Reinforcement Learning Applications Development: An Empirical Study
Mohammad Mehdi Morovati
Florian Tambon
Mina Taraghi
Amin Nikanjam
Foutse Khomh
2024-01-01
Empir. Softw. Eng. (published)
doi.org
arxiv.org
Data-access performance anti-patterns in data-intensive systems
Biruk Asmare Muse
Kawser Wazed Nafi
Foutse Khomh
Giuliano Antoniol
Data-intensive systems handle variable, high volume, and high-velocity data generated by human and digital devices. Like traditional softwar… (see more)e, data-intensive systems are prone to technical debts introduced to cope-up with the pressure of time and resource constraints on developers. Data-access is a critical component of data-intensive systems as it determines the overall performance and functionality of such systems. While data access technical debts are getting attention from the research community, technical debts affecting the performance, are not well investigated. Objective: Identify, categorize, and validate data access performance issues in the context of NoSQL-based and polyglot persistence data-intensive systems using qualitative study. Method: We collect issues from NoSQL-based and polyglot persistence open-source data-intensive systems and identify data access performance issues using inductive coding and build a taxonomy of the root causes. Then, we validate the perceived relevance of the newly identified performance issues using a developer survey.
2024-01-01
Empir. Softw. Eng. (published)
doi.org
arxiv.org