EECS 453 vs ECE/EECS 505 vs ECE/EECS 551 vs EECS 545 vs EECS 553

• EECS 453: Principles of Machine Learning
• ECE 505 (formerly EECS 505): Computational Data Science and Machine Learning
• ECE 551 (formerly EECS 551): Matrix Methods for Signal Processing, Data Analysis and Machine Learning
• EECS 545: Machine Learning (CSE)
• EECS 553: Machine Learning (ECE)

Important points about 505 vs 551

• ECE 505 and ECE 551 are very similar. They both use matrix methods extensively and both use the Julia language for technical computing.
  They are so similar that you cannot earn credit for both; you must pick one or the other. Do not enroll in both!
• For any course that requires ECE 551 as a prerequisite, ECE 505 will count as well, even if it is not officially listed that way.
  (We do not enforce prerequisites at the graduate level anyway, because they are just advisory.)
• ECE 551 is required for SIPML majors in ECE
• ECE 551 assumes that students have an EE background and already understand convolution and the DFT (e.g., from EECS 351).
• When enrollment is limited by room size, SIPML majors and other ECE students have priority for ECE 551.
• ECE 505 is targeted at a broader audience and usually is taught on central campus. Most non-EECS students should take ECE 505.
• If ECE 505 is not offered and you are unable to get into ECE 551, then another course to consider (for non-SIPML majors) is EECS 453: Principles of Machine Learning
• ECE 505 is an approved course for the MIDAS data science certificate. See listing here.
• There is a draft syllabus for ECE 505 for F20 at this url: [gdrive ]
  ECE 505 uses a cool "computational textbook" that Prof. Nadakuditi developed. You can learn more about it from his 2020 Juliacon video
• For comparison, here is a draft outline for ECE 551 in F20.
• If you are part of some other grad program that recommends ECE 551, then tell your adviser that ECE 505 is equivalent ask them to approve it for you. Have your adviser contact Prof. Fessler or Prof. Nadakuditi if they want to discuss it further.
• In recent years, ECE 505 has not been offered, so students should choose between 453 and 551.

Important points about 545 vs 553

• EECS 453 is targeted at senior undergraduates in EECS. Students outside the ECE program interested in machine learning are welcome as well.
• EECS 545 and EECS 553 are more advanced Machine Learning courses.
• There is quite a bit of overlap between 545 and 553, so you cannot take both for credit; you must choose one of them! Typically CSE students will choose 545 and ECE students will choose 553. We expect ECE SIPML majors to take 553.
• EECS 545 has "Coursework in probability, linear algebra, and programming" as advisory prerequisites.
• EECS 553 has "Graduate coursework in probability and linear algebra" as advisory prerequisites. Note that added word "Graduate" !
• In short, we strongly recommend that you take ECE 501 and ECE 505 or ECE 551 before taking EECS 553. The linear algebra background and the software experience from 505/551 are very helpful for 553.
• If you take EECS 445 first, then you *cannot* take EECS 545 for credit. However, you *can* take EECS 553 for credit, because EECS 553 builds more on the graduate background from ECE 501 and ECE 505/551.

Notes for UM ECE SUGS students in SIPML track

• ECE 501 and ECE 551 are both required for SIPML majors. They are both prereqs (formally or effectively) for many of the subsequent SP courses.
• 501 is offered both F/W.
• 551 is offered only in Fall (typically), so schedule planning is needed.
• 551 often has a long wait list, but be persistent because you have equal priority as any other SIPML graduate student and we will get you into it.
• 301 is sufficient preparation for 501.
• For 551, it is helpful to have some undergraduate linear algebra course first (as well as EECS 351).
• It is fine to take EECS 453 for your undergrad degree and then take EECS 553 for your graduate degree.

Notes