I dont understand why > Convolutional Neural Networks for Computer Vision is taken without traditional CV before, while NLP it says "include traditional NLP".

I dont understand why > Convolutional Neural Networks for Computer Vision is taken without traditional CV before, while NLP it says "include traditional NLP". I'd say it is a big nono to jump in to CNN directly without some basic CV, even-though CNN is outperforming most algorithm you will miss big fundamentals. Also most first/second year courses seems too CS focused, such as OS and Databases. I would remove those two things, add Differential Equation & Discrete mathematics (and numerical methods, maybe). At best the over focus of general CS stuff seems to create "CS with some ML experience" rather than a solid foundation of the principles behind it. This will give you a shoddy CS knowledge, and more fragile AI/ML understanding than someone with AI/ML background. I would call this program "Data Science" but not AI. Edit: removed "looks good" prefix after a second take at it.