My research has three main components. In information security, I work on the protection of privacy. Specifically, I am interested in the preservation of personal information that is spread over the Internet and accessed by services such as search engines, social networks, geolocation websites, online learning and e-commerce.

I use cryptographic protocols and different techniques for privacy protection: k-anonymity, randomization, secure multiparty computation and privacy by design. I also work to improve privacy policies concerning the categorization and confidentiality of sensitive data.

In e-commerce, I am interested in customization (acquisition of customer profiles) and recommendation of products and services using algorithms such as demographic, content-based, collaborative and hybrid filtering.

In the context of intelligent tutoring systems, I am interested in learning strategies, human-computer interaction, assessment methods and learner modelling. To do this, I use artificial intelligence techniques including machine learning and data mining.

My long-term goal is to understand intelligence; understanding its underlying principles would give us access to artificial intelligence (AI), and I believe that learning algorithms are essential in this quest. Learning algorithms could give computers the ability to capture operational knowledge (not necessarily in symbolic / verbal form) from examples.

Douglas Eck, research scientist and lead for Google Brain's Magenta project, is teaching computers how to generate their own music, video, images, and text using neural networks and other types of machine learning.

My research focuses on information retrieval and on Web search engines.

The goal is the improvement of the state of the art and the current practices in this field, through the development of novel information retrieval models, and by exploiting new data sources. These sources, such as user logs, Wikipedia entries and thesauri are put to use to expand, rewrite and otherwise reorganize user queries. My research interests also lie in taking into account the user’s various intentions in different application contexts.

To achieve this, we develop statistical methods to address the specific needs of information retrieval. My research also include multilingual aspects, i.e. successfully finding relevant documents in a language different from that of the query. The methods developed can be applied in various domains: medical information retrieval, e-commerce, opinion analysis on the Web, etc.

My research interests focus on the development of metaheuristics, such as tabu search and genetic algorithms, for solving discrete optimization problems in the transportation domain. I am particularly interested in vehicle routing problems with different side constraints, like service time windows at customer locations. These problems can model many real-world applications such as distribution of goods by commercial vehicles, courier services, para-transit services, etc. I also study dynamic variants of these problems when customer requests dynamically occur over time and must be integrated in real-time into the current routes.

Professor and associate member of MILA, the Montreal Institute for Learning Algorithms, Alain Tapp, is a long-time researcher in physics, computer science, and mathematics. From his early studies in these fields, he now brings his expertise to artificial intelligence and deep learning.

My research interests are centered around discovering fundamental computational principles that underlie the extraordinary capabilities to learn from the environment, understand it and adapt to it that characterize intelligence. The development of novel machine learning algorithms based on such principles, and trained on very large data sets, is at the heart of the latest technological breakthroughs in artificial intelligence.