Wednesday, December 17, 2008

graduate school

I'm happy to report that I survived my first semester in graduate school. It was a bit harder than expected, but I didn't realize how good of a school I am at until part way through the semester. As it turns out, Chapel Hill is a really, really good school. And so, I'm very happy to report that I have passed all of my classes.

On an even better note, I recently discovered that I am now doing exactly what I told myself I wanted to do 4 years ago. Four years ago, I took a class called discrete mathematics. I loved the class. The topics we learned in that class are still my favorite topics that I've ever learned. At that time, I told people and myself that I wished I could major in discrete mathematics, but unfortunately it appeared to me that there was nothing in discrete mathematics beyond that course. My school did not offer any classes on the topic, and I thought that was the end of it. I'm happy to report that most of the things I do in my new program are advanced topics of the things I did in discrete mathematics---graph theory, discrete time markov chains, and others. Furthermore, I've discovered a new pastime that I might spend some time with this winter break---binary programming. Maybe I'll write a program that can solve these hard Sudoku puzzles for me. Or maybe I can write a program to solve this impossible riddle I tried to do a few years back.

6 comments:

Jackijo said...

Can you tell me why UNC at Chapel Hill is a really, really good school?

Julianne said...

Apparently it's really hard to get into. Most people in my department received a perfect or near-perfect score on the analytical section of the GRE. I guess I did too... but I think all of the international students have a better work ethic than me. They're all really good students.

Jay said...

I'm glad you're doing well and enjoying it. You're going to have to explain what binary programming is. Also I see you mentioned markov chains. Are you doing anything with hidden markov models? From what I understand they work very well for temporal pattern matching (speech recognition, ocr, etc.).

Bradley said...

You never did tell us what that riddle was that you were trying to solve.

Julianne said...

It's a riddle called "Trouble in River City" which can be found here: http://www.twinbear.com/riddles.html. I'm still undecided if a linear program CAN solve this puzzle, but I'd like to try.

And, I've never studied a hidden Markov model, but I'd like to.

Ben said...

that riddle is too hard, it makes me want to do something fun instead of solving it...