CMU MSCF’s courses are terrible

QuantLearning

New Member
Only the math courses are so-so.
The ML courses and stats courses are just a joke..
Don’t be fooled by the course title.
They tried to cover all the popular topics, however they didn’t explain anything at all. You still know nothing when you completed the courses.
The assignments are poorly designed. No efficient feedback.

MSCF is the worst master program in CMU.
I took some popular courses from other departments, and the teacher’s overall teaching is much better.

This program’s only benefit is its career resource.
I highly recommend those who want do apply for MFE do not come to CMU MSCF. It’s just shit.
 

Michsund

Active Member
C++ Student
Only the math courses are so-so.
The ML courses and stats courses are just a joke..
Don’t be fooled by the course title.
They tried to cover all the popular topics, however they didn’t explain anything at all. You still know nothing when you completed the courses.
The assignments are poorly designed. No efficient feedback.

MSCF is the worst master program in CMU.
I took some popular courses from other departments, and the teacher’s overall teaching is much better.

This program’s only benefit is its career resource.
I highly recommend those who want do apply for MFE do not come to CMU MSCF. It’s just shit.
Can you give me your email so we could talk if your in the program currently? Just send me an email.
 

Michsund

Active Member
C++ Student
Only the math courses are so-so.
The ML courses and stats courses are just a joke..
Don’t be fooled by the course title.
They tried to cover all the popular topics, however they didn’t explain anything at all. You still know nothing when you completed the courses.
The assignments are poorly designed. No efficient feedback.

MSCF is the worst master program in CMU.
I took some popular courses from other departments, and the teacher’s overall teaching is much better.

This program’s only benefit is its career resource.
I highly recommend those who want do apply for MFE do not come to CMU MSCF. It’s just shit.
Also it seems that your are mad from a certain event that made you write something so negative. Do you really think cmu is that bad?
 

Onegin

Active Member
C++ Student
For the statistics courses, what in particular fell short from your point of view?

Your ML critique in particular is a bit jarring, as CMU are generally considered leaders in the field and the ML courses seemed to have positive feedback. Sounds like your experience was more of a "cookbook" approach, where you copy paste code without understanding the mechanism?

You might be on to something. It would be helpful to know more specifics.
 

imidig

New Member
Grammatical issues and obvious "emotional" bias aside... At most, you have taken 2 classes outside of the MSCF curriculum. You choose those two classes based on your preference and you have n=2... so idk how you overlooked observational bias as well as lack of data to reach the conclusion.

Let's get rigorous for the sake of argument. I talked to a few friends doing Msc in machine learning or computer science. Their entire 2 years focuses on 1 subject. Whereas at MSCF one needs to cover math, stats, cs, finance and communications in 1.5 years, and the program needs to leave time for students to interview for jobs. So if our machine learning class has not covered Markov decision process in detail, and somehow you are dying to learn more about reinforcement learning, then the right conclusion would be "I applied to the wrong program". It's so obvious that you are based in Pittsburgh not having the opportunity to talk to students from other MFE programs.

How funny this post is made after the final exams and the grades are coming out. I know for a fact that the over majority of your classmates would disagree with your conclusion. If you have an opinion, tell the program director. If you want to post information online, at least try to be objective.
 
So if our machine learning class has not covered Markov decision process in detail, and somehow you are dying to learn more about reinforcement learning, then the right conclusion would be "I applied to the wrong program". It's so obvious that you are based in Pittsburgh not having the opportunity to talk to students from other MFE programs.
This is why contrary to most of this forum, I don't think MSF or MFE degrees are a good idea.

How hard is it to get into an ML or CS masters program at CMU compared to their MSCF, do you have the stats?
 

Onegin

Active Member
C++ Student
@quantsmodelsbottles LMGTFY


Frequently Asked Questions
What are the prerequisites? Do I need an undergraduate degree in Computer Science? What test scores do I need?

Incoming students must have a strong background in computer science, including a solid understanding of complexity theory and good programming skills, as well as a good background in mathematics. Specifically, the first-year courses assume at least one year of college-level probability and statistics, as well as matrix algebra and multivariate calculus.
For our introductory ML course, there's a self-assessment test [PDF] which will give you some idea about the background we expect students to have (for the MS you're looking at the "modest requirements"). Generally, you need to have some reasonable programming skills, with experience in Matlab/R/scipy-numpy especially helpful, and Java and Python being more useful than C, and a solid math background, especially in probability/statistics, linear algebra, and matrix and tensor calculus.
The average scores of accepted applicants for Fall 2018 were as follows:
Undergraduate Overall GPA: 3.8 / 4.0, or 9.5 / 10.0, or 91 / 100.
GRE Quantitative: 169 (96th percentile)
GRE Verbal: 162 (88th percentile)
GRE Analytical Writing: 4.2 (68th percentile)
TOEFL: 111
 

Michsund

Active Member
C++ Student
This is why contrary to most of this forum, I don't think MSF or MFE degrees are a good idea.

How hard is it to get into an ML or CS masters program at CMU compared to their MSCF, do you have the stats?
You might feel like this as someone who already has offers from banks. Many people applying are using the degree to get there first job at a bank or hedge fund.
 
You might feel like this as someone who already has offers from banks. Many people applying are using the degree to get there first job at a bank or hedge fund.
it's a model validation internship role..not the buy-side hedge fund quant research (FULL TIME) stuff i ultimately want to end up doing. i'm not too keen on celebrating because i'm focused on the main prize, in fact i'm a bit depressed about it.

i don't see how an MFE is better at getting at a first job at a bank/HF than a CS/ML/Applied Math masters at a place like CMU.
 

Onegin

Active Member
C++ Student
it's a model validation internship role..not the buy-side hedge fund quant research (FULL TIME) stuff i ultimately want to end up doing. i'm not too keen on celebrating because i'm focused on the main prize, in fact i'm a bit depressed about it.

i don't see how an MFE is better at getting at a first job at a bank/HF than a CS/ML/Applied Math masters at a place like CMU.
It definitely used to be from my point of view, but it might not be anymore. You have a solid point.

I looked at data science programs, but they were less keen on taking old people like me. Also, I liked the idea of approaching ML from a finance perspective, rather than having the ML in my pocket, then trying to figure out how to apply it.
 

Michsund

Active Member
C++ Student
it's a model validation internship role..not the buy-side hedge fund quant research (FULL TIME) stuff i ultimately want to end up doing. i'm not too keen on celebrating because i'm focused on the main prize, in fact i'm a bit depressed about it.

i don't see how an MFE is better at getting at a first job at a bank/HF than a CS/ML/Applied Math masters at a place like CMU.
Place with cs usually get into tech roles. As for ML, these programs are rather new and I agree that you can get a degree in ML, but’s it’s gonna be extremely competitive to get in. Mfe degrees have recruiters coming to them so it makes getting a front office job much easier then from a math degree
 

Michsund

Active Member
C++ Student
Plus at cmu you can sit for phd ML courses as long as you can schedule it in, so it’s the best of both worlds.
 

Daniel Duffy

C++ author, trainer
It definitely used to be from my point of view, but it might not be anymore. You have a solid point.

I looked at data science programs, but they were less keen on taking old people like me. Also, I liked the idea of approaching ML from a finance perspective, rather than having the ML in my pocket, then trying to figure out how to apply it.
The challenge with new technologies is to show how they can be successfully applied to finance and in particular to say PDEs. Here is one attempt


It is not convincing to me for deep mathematical and numerical reasons. This is an area in which I just can't be hoodwinked, even if "Neural Network" is in the title.

The PDE community has similar doubts. The ML community are scared to answer our friendly-posed questions. Not even BS answers.
 
Top