Self-Optimizing Fabric for ENCQOR Network 5G- QC-328Discipline(s) souhaitée: Génie - informatique / électrique, Génie, Génie - autre, Mathématiques, Sciences mathématiques, Recherches opérationnelles
Durée du projet: Flexible
Preferred start date: As soon as possible.
Langue exigée: Flexible
Emplacement(s): Canada; Canada
Nombre de postes: 4
Établissements préférés: Université Concordia , École de technologie supérieure, École Polytechnique de Montréal, Institut national de la recherche scientifique, Polytechnique Montréal, TÉLUQ, Université de Montréal, Université de Sherbrooke, Université du Québec : Institut national de la recherche scientifique, Université du Québec à Chicoutimi, Université du Québec à Montréal, Université du Québec à Rimouski, Université du Québec à Trois-Rivières, Université du Québec en Abitibi-Témiscamingue, Université du Québec en Outaouais, Université INRS, Université Laval
Au sujet de l’entreprise:
Ciena is a network strategy and technology company known for its commitment to customer success. Our inspiration to innovate comes directly from our interactions that our customers have with our products and a commitment to stay ahead of our competition. Anchored by a best-in-class leadership team and approximately 5,700 highly-skilled professionals located in more than 80 countries, we support more than 1,300 of the world’s largest, most reliable networks.
Veuillez décrire le projet.:
This project consists in designing and implementing AI algorithms that will constitute the cognitive abilities of distributed collaborative agents embedded within these device swarms. The cognitive abilities of the agent community will create the intended collective and collaborative artificial intelligence of the system.
The algorithms to implement will include:
1) multi-agent algorithms to negotiate resources and coordinate decision making between agents operating within the same or different jurisdictions,
2) unsupervised and supervised learning algorithms to perform inference and extract insights from acquired data and predict future states/values of other agents and related components, and
3) reinforcement learning algorithms to derive optimal action policies for the agents.
The researcher will investigate algorithms related to his or her field of expertise.
Expertise ou compétences exigées:
We are looking for candidates with backgrounds in:
- deep reinforcement learning, to learn control strategies for individual agents, and possibly for coordinating multiple agents.
- deep learning & cloud computing, familiar with the lifecycle management process for machine/deep learning (data preparation, data labelling services, training, inference, deployment, model and data orchestration, etc.)
- multi-agent systems, familiar with multiagent architectures, parallel/distributed computing, networking, AI (the basics), and possibly game theory or IoT (a bonus). These two should also be excellent programmers.
- Masters in computer science (including AI)