Directory of Experts

You are here

Back to search results
FRQ-IVADO Chair on Software Quality Assurance for Machine Learning Applications

Phone: (514) 340-4711 Ext. 4233 Room: M-4213 Pavillons Lassonde
  Link(s)

Research areas description

Machine learning (ML) is increasingly deployed in large-scale and critical systems thanks to recent breakthroughs in deep learning and reinforcement learning. We are now using software applications powered by ML in critical aspects of our daily lives; from finance, energy, to health and transportation.However, ensuring the quality assurance of MLSA is still very challenging as evidenced by the recent deadly incident caused by the $47-million Michigan Integrated Data Automated System (MiDAS) or the Uber’s self-driving car that ran into a pedestrian even though the car’s sensors detected her presence.Traditionally, software systems are constructed deductively, by writing down the rules that govern the behavior of the system as program code. However, with ML, these rules are inferred from training data (i.e., they are generated inductively). This paradigm shift in application development makes it difficult to reason about the behavior of software systems with ML components, resulting in systems that are intrinsically challenging to test and verify. Compared with traditional software, the dimension and potential testing space of a MLSA is larger. Current existing software development techniques must be revisited and adapted to this new reality.

This Chair wants to research and contribute theories, methods, and tools to ease the development, testing, and release of high quality MLSA. Among the other contributions, the chair will identify good and bad development practices that can impede the maintenance, evolution and the reliability of MLSA. We will also develop techniques and tools to help developers detect and correct design defects in MLSA, both during design and implementation, and we will develop project management techniques adapted to the nature of MLSA. The chair wishes to develop efficient testing techniques and tools to help developers test the implementation of ML code, and to develop methods and tools to help document the behavior of ML models implemented in MLSA, with the aim to increase their interpretability and to ensure quality and reliability in these critical applications.

In addition to these research goals, the Chair aims to encourage diversity and equity in data science through the composition of his research team.

Research staff

Professors / researchers (3)
  • Foutse Khomh | Chairholder
  • Raman Md Saidur | Postdoctoral fellow
  • Houssem Ben Braiek | Master's student

External sources of funding

Fonds de recherche du Québec (FRQ)