Emin's Page

Grad School Advice

The following notes are from when I was asked to give some brief advice to incoming MIT grad students in 2003. The following notes may be copied, modified, and redistributed without royalty provided credit is given to the original author: Emin Martinian.

Advice For New Graduate Students

If you are in graduate school, you must have excelled at doing problem sets and passing exams. In the process, you probably became used to working by yourself on clearly defined problems and receiving regular praise from your professors. But in graduate school, the focus is more on research than on classes and the path to success is different.

One of the things I personally found most difficult to learn was accepting constructive criticism. In taking exams you become accustomed to solving the problem almost all the time and view the need for correction as a mark of failure. Research is different. In research, you fail more often than you succeed. A big part of doing good research is trying something, failing, getting feedback from your advisor and doing it better the next time.

Since you are probably only used to hearing good things from professors, accepting negative feedback may be difficult. You may be tempted to become defensive and argue when someone gives you advice. Try to keep in mind that your advisor is not trying to make you feel bad, but trying to help you do better. So make an effort not to take constructive criticism personally, but view it as an opportunity to improve. After all an important principle in engineering is that negative feedback is often much more useful than positive feedback: you often learn more from mistakes than from your successes.

Another strategy I found helpful was to learn as much as I could from older graduate students. Senior students often have all sorts of knowledge passed down through the years or learned the hard way. They can help you navigate university bureaucracy, figure out the computer system, find what papers to read, learn how to give a good presentation and many other things. To take one example, you will probably discover that reading and really understanding a paper is a long, difficult process. If you get someone who already understands a paper to explain it, you will save yourself a lot of frustration. In my experience, I learned as much from my fellow students as I did from my advisor.

Finally, a lot of your time in graduate school will be spent thinking about your own research. Unlike classes, nobody will assign you a clear set of tasks to accomplish in a set time frame. As a result, it is easy to get distracted, procrastinate, and never get engaged in doing research. This is dangerous. From the outside, a graduate student who is spending a lot of time thinking deeply trying to find a suitable research topic is difficult to distinguish from someone living idly. If you do get distracted and do not make progress for a couple of years, your advisor may not realize it until you have wasted significant time.

Do your best to set research goals and spend time working to achieve them. I found it helpful to keep notes on what I tried each week as well as on new ideas to try in the future. During a long dry stretch when I was stuck, I also kept a log of the papers I read along with a summary and analysis of each paper. Even though I did not make any direct progress for a semester or two, that process kept me engaged in research and let me hit the ground running when one of my ideas finally did work.

To summarize, being a graduate student is much different than being an undergraduate. To make the best use of your time, you need to adapt to these differences and not get stuck in "undergrad mode". I have mentioned a few things I wish I knew when I started. You may also find it useful to read some books about graduate school or look at advice people have posted on the web. One book I personally found useful was Getting What You Came for by R. Peters.

Good Luck.