This course provides engineering students with an in-depth understanding of generative artificial intelligence, focusing on the design, implementation, and deployment of advanced generative models. Students will explore foundational architectures such as Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), autoregressive models, diffusion models, and transformer-based systems like GPT. The course also introduces Retrieval-Augmented Generation (RAG), a powerful paradigm that enhances large language models by integrating external knowledge sources for grounded and context-aware generation. In addition to mastering core modeling techniques, students will examine recent trends such as foundation models, multimodal generation, and the integration of generative models within agentic AI systems (autonomous, goal-driven agents capable of reasoning, planning, and tool use). Hands-on projects will allow students to apply these concepts to real-world tasks involving text, image, audio, and cross-modal generation. Ethical considerations, including bias, misinformation, and responsible deployment will also be discussed. By the end of the course, students will be prepared to build, fine-tune, and evaluate generative AI systems in both industrial and research contexts.

Temps présentiel : 30 heures

Charge de travail étudiant : 70 heures

Méthode(s) d'évaluation : Examen final, Examen partiel, Travail personnel