COM 428 GENERATIVE DATA MODELING
Generative models have gained popularity and reached common use in the public space, being able to pass the bar exam and complete first-year programming assignments with ease. But underneath the hype and wonder, lies a carefully structured scientific methodology that has its foundation in probability distributions. In this course students will explore how data can be used to construct plausible generators along with careful discussions on limitations, bias, and the ethical issues that automated content-creation gives rise to. Open-source models will be used in the Python programming language to develop applications that make use of pre-trained models, while further exploration will be encouraged in a culminating course project.
Prerequisite
Either
COM 304, 307, or 316. COM 328 is suggested.
Enrollment Limit
Enrollment limited to 18 students.
Attributes
MOIC