Course: COMP 646. Generative AI (3)
Prerequisites: COMP 442 or COMP 542 or COMP 643; MATH 340 or equivalent; or instructor’s consent. Recommended Preparatory: COMP 643. Generative artificial intelligence (AI), a rapidly advancing subfield of machine learning, focuses on creating models capable of generating data such as images, text, and audio. This course covers mathematical and computational foundations of generative modeling, as well as applications in engineering, design, science, and the arts. Specific topics include generative adversarial network (GANs), autoencoders (AE), variational autoencoders (VAEs), transformer-based model, diffusion model, autoregressive model, information lattice learning (ILL), normalized flow model, neural text decoding and prompt programming. It will also explore ethical considerations, societal impacts, and the implications of AI-generated content, with a focus on issues such as intellectual property, authenticity, bias, and safety. This course is designed to provide graduate-level students with a comprehensive understanding of the methodologies, technologies, mathematical foundations, and algorithms essential for applying generative AI to a wide range of applications.