Updated MSBA Curriculum: Stronger Focus on AI

Ashwin Aravindakshan Nair, Associate Professor and MSBA Academic Director
So what I usually do when I come here, I introduce myself, of course, and I'll do that in a minute. But I talk about the curriculum quite a bit. And what I'm going to spend time on today is not just talk about a curriculum, but talk about the changes that are happening to a curriculum and how.

We have been working on this for about a year. And I'd say probably about a month ago, we got full approval to roll this curriculum out. So that's a very exciting development in our program, which I'll get to in a minute. So for those of you, you know, I've spoken to a few of you, but for those of you whom I haven't spoken to, my name is Ashwin Nair.

I've been at Davis since 2007. The work I do is, you can see at the intersection of, you know, behavior and analytics and machine learning. And so I'm sure you might have heard of these, custom analytic models, developing customer lifetime value models, cohort analysis and so on, and how advertising might impact there. That's the kind of research I do.

But I also do research on the causal side of things. So trying to understand or explain why something happened, and using data to try and... come to those conclusions. So that's really my research focus. And in the program, when I used to teach, I used to teach what you might think of as the causal analytics course, where we dive into understanding how data, the methods that can be used to try and point that, yes, you, the firm doing this improved ROI by this much, and hence we can develop some sort of risk of analytics based on that, and so on.

Now, moving on to the curriculum, as you might have seen in industry, there is massive changes happening in the industry with the advent of chat GPT, what is it now? A year and two months ago, three months ago. So it's not that long. And all the other new gen AI models, the utilization of generative AI in industry has grown dramatically.

So what does that mean in terms of your skills? Well, your skills now need to encompass not just analytics and machine learning, but also need to encompass, also need to include knowledge of generative AI, the modeling of generative AI, the training of generative AI, the use of things like LORAs and QLORAs and so on and so forth that have become. industry standard today, right? At least the cutting edge industry standards. So not all firms use it.

And so what we wanted to do in the program is to try and bring this in and incorporate it into our courses. So there are specific courses that are built specifically for artificial intelligence. And so where you, one of the course, for example, I'll get into it in a minute, you will actually design an LLM, right? So you won't go to GPT-4, we don't have the GPU resources for that, but you will probably do something similar to a GPT-2 type model.

So, you know the bare bones of what it takes to build one of these and train one of these and so on. And so all that is being incorporated into the program. But apart from that, we have now incorporated AI into multiple courses. And I'm sure when Professor Beam talks about her courses and her program with the practicum, she'll talk about the focus that is there now on using what you might think of as AI assisted analytics in the, you know, as you work through your practicum or as you enhance your learning.

So the first point to, you know, with this curriculum change is that there's a lot of focus on not just business analytics, but the incorporation of artificial intelligence into every aspect of the courses. Second,

Before our program was a 10-month program, right? And we did that with a specific intention.

Now, with the incorporation of AI and the importance that we feel, you know, AI is not simple. If it was simple, everybody would do it well. So for you to actually become, I wouldn't say, you know, you will become a world-class expert by the end of this program, because that takes a PhD.

But to become an expert that industry needs by the end, you actually need more time to work with these models. And so what we have done is rather than have a 10-month program, we now have a longer program and we get to that. So you start in summer one and you will end not in June, but at the end of the second summer. And this allows you to work more with models and there are some other advantages that you'll get to in a minute.

And so you can think of this as a slightly longer program by about two - three months. The next point is that because this becomes a longer program, you will have more time to job search, to look for internships, and to also hone your skills better, right?

So, before everything had to be done within 10 months. Now, because you have the second summer, you have a lot more time to work on your applications. to build your resume out, to get internships and any other trainings that you might need and so on.

The other thing that we incorporated this time is the option for more electives. So before we had only one elective, but now we are, the way we look at electives is, we want you to develop the ability to go from zero to pretty well worst in a domain. very quickly, that's our goal.

Because by the time you get to the electives, we know that you are skilled enough to build all these machine learning models that we usually teach in the program. But one thing that we have heard from industry quite a bit is that students, and it's not our students, one of the reasons our students get placed well is because we have this focus.

But a lot of times when students come in, they have a very hard time understanding business. And that's really the focus of these electives, to get you from zero to understanding pretty well a business within a short span of time. And so we are now including an extra domain course so that you get one more opportunity to learn a new domain and you develop the skills or strengthen your skills to learn about domains very quickly.

And the other thing that we have included this year, which is completely new, is a research option. Over the past few years, we have had several students who wanted to work with faculty on research, but we never had a way for them to actually do a thesis.

There were several students who said that they would be interested in, for example, working on a new problem that the professors are working on, and perhaps take a slice of that and make it their thesis, but they come up with their own idea. The thing with the thesis is you have to come up with your own idea.

And so the research option that we have in allows you to add an extra, I'd say maybe four months to your program. So beyond summer, where you work on nothing but a thesis, a master's thesis. And so this is different from an internship.

This is just you focusing on an advanced research problem that you have proposed and you intend to solve with guidance from a professor.

So what does this look like when we think about the comparison? So in terms of credit units, if you take the standard MSBA program, it's the same. So it's 44 units each. The span of the program used to be early August to mid-June. Now the program goes from early August to late July, but your actual official graduation date is in September. I think around mid-September usually.

Elective coursework, it used to be one. Now it's two. In terms of credits per term, in the first summer, you get six units. Fall is 12, winter is 12, spring is 12, and then the second summer, it's two units again. If you think about an internship option, there was no internship option before. Now you do have a 20-hour, six week internship option in summer two. And then of course, you know, that time that you get in summer too, as you can see, you know, your coursework basically gets done by mid to mid-July to late July. So summer two, those goes on for another month, month and a half. And so that extra time that you have, you can spend solely focused on job searches, internships, and so on and so forth, or strengthening your skills. And then we have a two term option for a research extension, but that requires an extra six credit units.

And if you're interested in that, I'm happy to talk to you. If you can answer a few questions now, but if you join the program, then that's something we can always discuss in more detail. 

So in terms of your courses, what does this look like? In the first summer, what we do is we build a ramp. so that we know by the end of summer, all students are on the same level playing field. Because one of the things we found very early in our program is that is not always the case. Because students come with different backgrounds and even though maybe you try and learn on your own, it's not the same as being guided or given curated content that our professors put together.

And so in the first summer, the foundations course is essentially building you up so that you're ready for the major coursework that comes in fall. The first summer also is when, and I won't talk a lot about practicum because Professor Beam is gonna talk about it, but I'll just say a little bit about it.

The first summer is when you get introduced to your practicum projects, right? And that's when I presume teams will form and you'll start getting some exposure to the company and who your main contact with the company is and so on. Then in Fall, You can see through color coding, you get introduced to the different bins that are there in business analytics.

So for example, you get exposed to statistical reasoning and exploration, which is not just your introductory stats course, but it's a fairly advanced, what I would say, you might think of as an intermediate level stats course, because the foundations is where you learn your basics.

Then you get introduced to data management, data engineering principles and so on. And then in information insights and impact is when you learn things about problem structuring, A-B testing, experimentation, causal analysis and so on. Then you have the practicum again. which then flows throughout the year.

In the winter, we get to the more advanced level in statistics where you're introduced to concepts around time series modeling, filtering techniques, and so on. You also, as you can see now, I haven't really, you know, mentioned it because I mentioned it earlier. But in each of these courses, we have incorporated principles of artificial intelligence. In the machine learning course, though, is where you dive into the brick and mortar of artificial intelligence, where you start building these LLM models. You learn about neural nets, how they work.

You get into deep neural networks, and then you start building LLMs. So that by the end of machine learning artificial intelligence, you are very well versed with how these models work and how to put them into practice. Okay. And then we have another course that further enhances your data engineering skills called data design and representation where you deal with unstructured data, you deal with streaming data and so on. And so you work with more common, less well-defined data structures. And then you move to spring where we have a course called big data, but big data is a combination of multiple things that you've done in the past.

So for example, big data employs your data engineering skills that you have learned till now, your machine learning and AI skills and your statistics skills and adds to them by introducing things around concepts around distributed computing. So you know, if you use PySpark or some other form of How do I allocate resources across multiple GPUs? So all those kinds of questions are what we explore and try to answer in the big data course. And then you have analytic decision making, which gets into more of the prescriptive analytics type knowledge.

So you might think about if I want to optimize my, so if I want to maximize profit, what is the optimal spend on advertising. This is a very simple problem that I've just laid out, but if you really think about the problem structured correctly, it becomes extremely complex very quickly. And solving such a thing requires many advanced methodologies, which is what you'll be working on in that course.

And the reason that course isn't the last part is because it requires multiple courses from the previous quarters so that you're prepared for that. for that course. And then we have multiple electives. We listed some of them here, people analytics, application domains, so logistics and supply chain, but we'll be introducing newer ones for next year also.

And then comes the final summer where we, you will work with Professor Beam and Professor Martinez on basically implementing your practicum, right? So in a nutshell, that's our program. As you can see, it's a mix of the technical side, that's the computing and data.

So one of the reasons why our graduates are so highly sought after is precisely because of the mix of these. So it's not that, you know, They understand business, but they don't know how to implement. It's that you know how to implement, you understand what it takes to actually put these models into production, and you can also define the impact that these models will have on business.

So you can tell the entire narrative of your model or the efforts that you will employ when you put these models into production. 

Question from Zoom participant
So is the number of electives we can choose at max is two?

Professor Ashwin A. Nair
You can always take more, but we'll count two. Towards the degree, you only need two. But there are students who are wanting to take electives in the MBA program and that's fine. The thing is every course needs to be approved for the MSBA program because, you know, we promise companies, for example, a certain level of technical competence. And so in that sense, yes, you can take more electives. It's just that there will be some set of approved electives that can be counted. And there are exceptions also, but at that time you want to speak to the academic director about it.

Question from Zoom participant
Are there companies that we could choose for our internship?

Eric Duarte-McDermott, MSBA Executive Director
So you will work with our career development team to be able to help identify the companies that are looking for interns and we'll help prep you along the way. However, we always say that we are not here. It's a partnership. So, you know, you put in your work, we'll help you along the way. So it's 50-50 and you identify the companies, we'll help you identify the companies, you're the ones who are gonna be applying and we'll help prep you for the interviews and your resume.

And so last question we will take before we kind of, before we move on.

Question from Zoom participant
At present, tech job market is very poor. Many of our seniors are not able to get jobs within three months after completion of the course. if I wish to enhance my course duration by four months after one year as a research assistant with professor?

So you won't be a research assistant because you're working on your own thesis. So that's why this is different. This is six credit units that you'll be taking. And this is specifically if you want the thesis option. One of the key advantages of this lengthening of the program is that our coursework basically ends by mid July. but you don't officially graduate till mid September.

So you have a long span of time built in already where your initial, what is that clock called? I forget, Eric. The OPT. Yeah, so there's an OPT clock. Yeah, and that won't start till you finish, I think.

Eric Duarte-McDermott, MSBA Executive Director
Perfect. And I'll take two more questions because the one is a fee structure question. So the fee structure doesn't change. It's still 44 units long. However, you have to account for longer cost as far as living costs. So as far as tuition goes, it's the same 44 units. Our tuition is going to be sixty three thousand one hundred and forty dollars this year. And but like I said, it's still the 44 units that was similar to that was identical to our last unit structure

And then so I'll take this last question because I think it's a good question for me to answer and a great segue into Professor Beam's.

Question from Zoom participant
So during the second summer, we will finish our practicum implementation. And when would the option for the internship be?

Professor Ashwin A. Nair
Great question. I will let Professor Beam talk about the practicum implementation because she's going to speak next about this, but you would take it in the summer too. So when you're finishing your, if you choose to, so there's not an internship that's a requirement, but it's a two unit course that you'll be taking in the second summer session that will allow our students to be able to take a 20-hour internship if they choose. So that's where the six week, 20-hour internship goes. It will, it will happen during your second summer, right, as you finish up the program.