How to succeed in this course

For many students, this is your first college-level course in mathematics. In high school, you may have taken calculus courses that taught you to perform (sometimes rather involved) mechanical procedures for computing derivatives, integrals, and the like. About half of this course will be similar, except instead of computing integrals, you'll learn to solve systems of linear equations of various forms.

The other half of this course, however, will likely be more abstract in nature than anything you've seen before. You are at the Georgia Institute of Technology, after all — this is not high school any more. We will focus on conceptual ways of understanding equations and their solution sets. We will ask questions like, what is the dimension of a set of solutions? What are all ways we can write them down? Are there any properties of the matrices involved that we can exploit to describe the solution sets, without actually solving the equations in the first place? What kind of geometric questions can we associate to a system of linear equations?

For these reasons, the methods you used to approach your previous mathematics courses may not suffice any more. For instance:

  • You will probably notice that there are lots of vocabulary words in this course, such as span, linearly independent, eigenvector, invertible, etc. You have to not only learn what they mean, but learn to make them work for you. The language of mathematics is very powerful, but also very precise; having a vague idea of what a term signifies is not useful.

  • You may be tempted to ask, "so what's the recipe for solving this kind of problem?" or "what kinds of problems will be on the exam?" About half of the problems will not reduce to some kind of mechanical procedure, so there really is no recipe. I can't tell you that you'll only have problems of a certain form, and that I'll just fiddle with the numbers on the exams. If Wolfram Alpha can do the problem faster than you can, it doesn't count as problem solving. Problem solving is the part where you figure out what to ask Wolfram Alpha (or Mathematica, etc.) in the first place. And this is as it should be: nobody is going to hire you to do calculations that a computer can do better.

  • If you've read this far, then you may be worried that your usual approach is useless, and that you're in for a tough semester. Don't worry, you're smart and hardworking. You wouldn't have been admitted if we didn't think you'd succeed. You'll figure it out! Plus there are lots of resources available, so you won't be on your own.

Read on for more specific advice.

  • Do the homework assignments carefully
    This is the only graded portion of the course where you have a full week to think about your answers. The only way I know to learn to do math is by practicing doing math — just like the only way to learn to play tennis is by spending time on the practice court. The sports analogy is quite good — it amazes me that it's obvious to people that pratice is important to do well in sports, but not in math.

  • Keep up with the daily coursework
    Trying to learn everything the night or the week before an exam is like trying to learn to hit serves and backhands in one long day of practice with a big match the next day.

  • Work with your classmates
    In addition to being a great way to get to know people, you'll find you may learn as much from your classmates as you do from your professors. I'm of the firm opinion that the advantage of being at a prestigious institution such as Georgia Tech is as much the quality of your peers as the quality of instruction.

  • Read the textbook before the lecture
    The purpose of lecture time is to fill the gaps in your understanding, not to communicate all of the content of the course. You don't need a Georgia Tech professor to teach you the easy stuff, so I'll concentrate on the tricky points.

  • Come to class and to recitation, and pay attention
    I prepare lectures carefully, and the online lecture notes will not mean much to you unless you were there to hear them explained. If you're not in class, then you won't hear me say the word "span" four hundred times in one week, and therefore you might be surprised to see it on the midterm. Besides, attendance is part of your grade.

  • Take advantage of the resources provided
    We know that learning linear algebra can be hard. There are lots of ways to get help. Come to office hours! Read the textbook! Go to Math Lab! Ask questions on Piazza! We want you to succeed, but you can lead a horse to water...

  • Don't look for answers on the Internet
    This is not because it's considered dishonest; rather, it is hard to find quality information on the Internet, and even harder to judge the reliability of any information you do find. Why spend your time wading through sketchy YouTube videos and Stack Overflow answers when you have an extremely carefully prepared and curated set of resources right here on this page (for which you have already paid a great deal of money)? At the very least, only use the Internet as a last resort, after you've already been stuck for a while.

Resources

  • My office hours
    These are set periods of time where I sit in my office (Skiles 221), staring at the door and hoping my students will show up to ask me questions. Seriously, I chose to be a college professor in part because I honestly enjoy interacting with students, and I derive a lot of satisfaction when they end up learning some math. That said, do come prepared with questions that you've already thought about a bit. Don't expect me simply to tell you how to do the problems; this is not a good way to learn.

  • The TAs' office hours
    Your teaching assistants are also knowledgeable about linear algebra, and have scheduled time to answer your questions. Whereas you must attend your assigned section, you are free to visit the office hours of any of your TAs.

  • Math Lab
    This is essentially a free walk-in one-on-many tutoring service provided by the School of Mathematics, and is staffed by graduate teaching assistants. Its hours are very flexible. Here you will find people who are willing to spend lots of time helping you and answering your questions. Note that it's not necessary to find a time when you can meet with a graduate instructor who is currently teaching 1553: any math grad student is very comfortable with basic linear algebra.

  • The textbook and course materials
    If you get stuck, Go back and actually read through the textbook carefully. I promise, everything is in there, or in supplemental material available on the calendar. Other course materials, such as lecture notes, solved worksheets, quizzes, exams, and practice exams are also located on the calendar. I'll post new material throughout the semester.

  • The exercises in the textbook
    There are many good exercises at the end of every section and every chapter of Lay. These are ideal for use as practice problems.

  • Piazza
    This is an online space for this class (1553-B*) where you can post questions and help other people by answering their questions. I will also check Piazza regularly and respond to questions. It can be accessed through the course site on T-Square. See the discussion on the organization page. Please login to it at least once, since you will not receive class announcements otherwise.

  • One-on-one tutoring
    If you decide you need extra, personalized help beyond what is freely available, you can consider hiring a tutor.