?

Log in

No account? Create an account

Previous Entry | Next Entry

Goodbye to you

Smeagol voice, redux:
Weeeeee're freeeeeee!

CIS 690 wrapped up today, to not much fanfare.
Projects look all right this year. Most are even a little promising, though I've only had one complete project interview out of five at this point.

I also said goodbye to my sixth M.S. thesis advisee1 this afternoon. I found myself shaking his hand, when he suddenly realized that that afternoon would be the last time we met before he went back to the Slovak Republic. We probably won't see each other again for quite a few years.

Is it weird that I feel a little nostalgic and a little sad at the departure of students when they finish up?
I should be happy for him, no? He's defended. He's graduating.

Somehow, it doesn't make it easier, especially in the case of overseas students or those going overseas. Experience doesn't make it easier, either. Odd, that.

1 He is my 6th M.S. student but the 8th to complete an M.S. program under my supervision, counting a student (Agah, 2001) who had an M.S. project and another (Valluri, 2002) with a coursework only ("nonthesis, nonreport") option. This is not counting Master of Software Engineering students, of whom I've had six.

--
Banazir

Comments

( 22 comments — Leave a comment )
myng_rabbyt
Jul. 7th, 2004 06:31 am (UTC)
I think it's lovely that you care. I don't think it's weird at all that you feel a little nostalgic and little sad. These are people you've taught and helped and encouraged and nurtured intellectually. It's like little birds leaving the nest. Of course you're going to be sad. And, sorry to sound so cheesy, I think it's great you care about them.

I knew there was more than one reason I liked you, Banazir.
prolog
Jul. 7th, 2004 07:00 am (UTC)
Agreed. If you've worked with this person for the last couple of years, why shouldn't you feel that way?
banazir
Jul. 7th, 2004 07:55 am (UTC)
Thank you
Agreed. If you've worked with this person for the last couple of years, why shouldn't you feel that way?
Well, just one year, but thanks.
Your advice is somthing to think about.

--
Banazir
prolog
Jul. 7th, 2004 08:17 am (UTC)
Re: Thank you
He finished his master's in one year?

Hardcore.

(here, at least, pretty much the earliest I've seen done is 16 months - the class component means that one's thesis component is delayed at least eight months)
banazir
Jul. 7th, 2004 09:04 am (UTC)
Hardcore indeed
He finished his master's in one year?
Hardcore.

Yep.
Intro AI and Machine Learning last fall, Advanced AI this spring. He started writing the thesis about four months ago and has been running experiments nonstop on the aforementioned quad Opteron cluster.

(here, at least, pretty much the earliest I've seen done is 16 months - the class component means that one's thesis component is delayed at least eight months)
Wow.
Did I mention that I tend to attract hyper students? :-D
Hyper-minded, anyhow.

My M.S. project took roughly two years (summer 1991 - summer 1993), but it didn't really count towards my coursework-only Master of Science in Engineering (M.S.E.) degree at Hopkins. Also, I was in a concurrent 4-year B.S./M.S. program (well, 5-year, but yeah, I like pain, kindasorta).

--
Banazir
prolog
Jul. 7th, 2004 09:17 am (UTC)
Re: Hardcore indeed
Wow. Does he have a girlfriend? That tends to influence how much time one can spend on schoolwork. ;)

If he does have a girlfriend, she must be inordinately patient!

Here, M.Sc. students have to take five courses, plus attend at least a dozen seminars over the course of their degree. I've still got one to go, because I didn't do five-in-one-year like some do. Then again, I wasn't going bat crazy in term two, when those that take that path have to do three grad courses (ouch!).

My current plan is to have my thesis proposal ready for early September, and work on it concurrently in first term while taking a course in Intelligent Systems, then to finish it up in second term. I need to go hard because there's a gathering of my dad's side of the family in England next summer, and I'll be damned if I miss that!

That reminds me - I need to get around to registering for next year. No rush, but still, I keep forgetting!
prolog
Jul. 7th, 2004 09:18 am (UTC)
Re: Hardcore indeed
Err, by "one to go", I meant classes, not seminar. Grammar eludes me when I wake up at 6:00 a.m.
banazir
Jul. 7th, 2004 10:05 am (UTC)
The Grammar Hunters
Err, by "one to go", I meant classes, not seminar.
Grammar eludes me when I wake up at 6:00 a.m.

Hrm, grammar eludes me, too, but I just hire bounty hunters to tear my lab apart until they find those elements of grammar and style. (Grammaticons?)

No disintegrations!

--
Banazir
prolog
Jul. 7th, 2004 12:09 pm (UTC)
Re: The Grammar Hunters
As long as the Grammaticons are led by Megatron, I'm all in favour of this.
banazir
Jul. 7th, 2004 06:38 pm (UTC)
Re: The Grammar Hunters
As long as the Grammaticons are led by Megatron,
I'm all in favour of this.

... 'til the day we are one?!
One World Order?!?!
I'ma tellin' zaimoni!!!

--
Banazir
banazir
Jul. 7th, 2004 09:47 am (UTC)
Tales out of school
Wow. Does he have a girlfriend?
That tends to influence how much time one can spend on schoolwork. ;)

Hrm, no idea. If he does, she might be back home.
I've had students whose SO's came out to see them, though.

I know I didn't during the time I was plowing through my M.S. program. (Long story, very sad; hempknight knows it.)

If he does have a girlfriend, she must be inordinately patient!
Duuuude, you said it.
I'm always so very pleased to hear deans take note of the large number of bachelors in our department, and to hear my senior colleagues in the field explain how computer science is a selfish discipline. (It is true, though.) :-P

Here, M.Sc. students have to take five courses, plus attend at least a dozen seminars over the course of their degree.
Hrm, I had to take ten (30 semester hours).
No seminars, no thesis, though.
I just did the work of a thesis - served me well later on in my dissertation, though.

I've still got one to go, because I didn't do five-in-one-year like some do. Then again, I wasn't going bat crazy in term two, when those that take that path have to do three grad courses (ouch!).
Three? Three?
Any of you fine UIUC folks remember Albert Tsai dropping from 5 to 4 during our first semester? He looked half-dead on his feet.
Mighty Ramachandran took 5 and did not know he could drop until his academic advisor told him... halfway through the semester.
My academic advisor, Geneva Belford, was rather more experienced and told me to take no more than 3 (full units, so 12 graduate semester hours) with no more than 2 heavy programming courses during my first semester in 1993.

My current plan is to have my thesis proposal ready for early September, and work on it concurrently in first term while taking a course in Intelligent Systems, then to finish it up in second term.
Cool. Keep us posted!

I need to go hard because there's a gathering of my dad's side of the family in England next summer, and I'll be damned if I miss that!
Get a paper into IJCAI-2005 or a satellite workshop thereof, maybe?

That reminds me - I need to get around to registering for next year. No rush, but still, I keep forgetting!
:g:
Want an auto-reminder?
I still have the kdd-reminder Yahoo! Group, aka the MassPHAr (MASSively Parallel Harassment ARray).

--
Banazir
hempknight
Jul. 7th, 2004 11:01 am (UTC)
Re: Tales out of school
I know I didn't during the time I was plowing through my M.S. program. (Long story, very sad; hempknight knows it.)

Que? I know noooooooooooothin'! I'm from...Barcelona...

Everything I think I know, you think I think you think I think I know is something I have no recollection of and am willing to take you to court of law to prove that I have no recollection of any events of such nature be it past, present or future.

So there!

--
Danger is my middle name
banazir
Jul. 7th, 2004 08:02 pm (UTC)
Re: Tales out of school
Que? I know noooooooooooothin'! I'm from...Barcelona...
Bha.
Go tell nobuddy69 a story.

Everything I think I know, you think I think you think I think I know is something I have no recollection of and am willing to take you to court of law to prove that I have no recollection of any events of such nature be it past, present or future.
Lies, lies, all lies!

So there!
Nyah.

Seriously, you dknot memember?

--
Banazir
hempknight
Jul. 8th, 2004 12:12 pm (UTC)
Re: Tales out of school
Seriously, you dknot memember?

Refresh my memory...eh? *holds hand out*

;)

--
Danger is my middle name
banazir
Jul. 8th, 2004 04:50 pm (UTC)
Refresher
Refresh my memory...eh? *holds hand out*
;)

*pays you a crystal flokarino*

"If she's not on board within seven hours,
My train's leaving the station,
With or without her."
- a TEUNC who is knot moi

--
Banazir
prolog
Jul. 7th, 2004 12:33 pm (UTC)
Re: Tales out of school
I'm always so very pleased to hear deans take note of the large number of bachelors in our department, and to hear my senior colleagues in the field explain how computer science is a selfish discipline. (It is true, though.) :-P

It really is. I've yet to meet a subject that equals the CS workload. I've heard engineering does, but not being an engineer, I wouldn't know. It's a lot of work, but it's the discipline I love, so I don't really mind. My girlfriend has mentioned that she wouldn't mind seeing more of me, though. Heh.

Hrm, I had to take ten (30 semester hours).
No seminars, no thesis, though.


Sounds about right. If you budget for five courses per year, and don't do a thesis, it sounds like an equivalent amount of work.

Is that common at American universities? A friend of mine was accepted to Stanford for Mechanical Engineering (but had to turn them down because he couldn't get enough funding), and he said that the program there was pure coursework, too.

My academic advisor, Geneva Belford, was rather more experienced and told me to take no more than 3 (full units, so 12 graduate semester hours) with no more than 2 heavy programming courses during my first semester in 1993.

That's basically how it works here, too. However, since I took pretty much all of our theory classes (complexity, compilers, formal AI, etc) as an undergrad, the stuff left to me as a grad student was...programming. And since I'm not going to take three programming-intensive courses in one term (well, I could, but I didn't really want to break up with my girlfriend), I did it this way. :)

And I should point out the ones going crazy took three programming-heavy courses. A friend of mine, who took a couple of theory courses and one on software engineering, was fine. As you point out, it's all about the workload of the individual classes.

Cool. Keep us posted!

Will do. At the moment I'm just working on my thesis proposal, and looking into places where I might be able to do a Ph.D.

Get a paper into IJCAI-2005 or a satellite workshop thereof, maybe?

Might be possible, but unlikely. IJCAI is mostly more formal AI, isn't it? My stuff is pretty applied - modelling, service selection, classification, etc. Is that what the workshops are for? (I don't know much about the whole conference process - I submitted my first paper about a month ago, and don't get word until July 23)

Technically, I have some unfinished research in a completely unrelated area that I could submit (on Bayesian approaches to phase transitions in NP-hard problems) that was my undergrad thesis, but it's very unpolished, and probably needs way more experimentation than we did.

Want an auto-reminder?

Nah, that's okay. The cool thing about grad studies here is that registration has a far greater window than undergraduate studies. When I was an undergrad, if you weren't on the phone registration system the moment you were eligible, most of the classes would be filled. If the class I want to take is filled, I can just go down the hallway and get Dr. Vassileva's signature. :)
banazir
Jul. 8th, 2004 05:53 pm (UTC)
Conferences, Part 1 of 2
I'm always so very pleased to hear... my senior colleagues in the field explain how computer science is a selfish discipline. (It is true, though.)
It really is. I've yet to meet a subject that equals the CS workload.
It's not just workload; it's mindset.
Impatience for results, believing the world began in 1971, etc.
What was it about impatience, arrogance, and hubris being programmer virtues? (I don't think they are, and in fact I feel a rant coming on about it, but I'll save it for another entry. :-))

I've heard engineering does, but not being an engineer, I wouldn't know.
CS is in engineering here, so insofar as I consider myself an engineer (only to a limited degree): it's true.

It's a lot of work, but it's the discipline I love, so I don't really mind.
I don't mind the workload or the hours. I do mind constant interruptions after hours of time spent with my friends, but I've gotten somewhat used to it. Not entirely, mind you.

My girlfriend has mentioned that she wouldn't mind seeing more of me, though. Heh.
Glad to hear it! As I hope you are. :-)

Hrm, I had to take ten (30 semester hours).
No seminars, no thesis, though.

Sounds about right. If you budget for five courses per year, and don't do a thesis, it sounds like an equivalent amount of work.
That's a pretty rich thesis.
Ours require a minimum of 6 hours of CIS 899 (thesis credit).

Is that common at American universities? A friend of mine was accepted to Stanford for Mechanical Engineering (but had to turn them down because he couldn't get enough funding), and he said that the program there was pure coursework, too.
I don't know much about the grad programs of universities other than Johns Hopkins, the University of Illinois, and Kansas State University. It sounds plausible, though.

My academic advisor, Geneva Belford, was rather more experienced and told me to take no more than 3 (full units, so 12 graduate semester hours) with no more than 2 heavy programming courses during my first semester in 1993.
That's basically how it works here, too. However, since I took pretty much all of our theory classes (complexity, compilers, formal AI, etc) as an undergrad, the stuff left to me as a grad student was...programming.
Oh, I see. I have some very good theoretical CS students who put off their systems and development courses (OS, Software Engineering capstone, etc.) as long as possible, and would have deferred their grad ones too, if we hadn't pushed them. I personally tried to even them out. What specific courses did you have to go through first as a grad?

And since I'm not going to take three programming-intensive courses in one term (well, I could, but I didn't really want to break up with my girlfriend), I did it this way. :)
Good choice. :-)

And I should point out the ones going crazy took three programming-heavy courses. A friend of mine, who took a couple of theory courses and one on software engineering, was fine. As you point out, it's all about the workload of the individual classes.
Yes, it is. Some are downright notorious.
I always avoided taking three programming-heavy courses.
As an undergrad, I did it, but:
a) I was 16-18 and rather full of teunce;
b) I was a fairly good (C/C++) programmer at the time;
c) I fudged it by taking 1 theoretical CS, 1 math, and 2 programming-intensive courses (12 hours). The fifth course was usually an independent study I had already done the work for over the summer (3 additional hours), and I would just ask my professors to let me count it in the the fall.

Cool. Keep us posted!
Will do. At the moment I'm just working on my thesis proposal, and looking into places where I might be able to do a Ph.D.
Please tell me again: what program areas are you interested in?

(continued)
prolog
Jul. 9th, 2004 02:50 pm (UTC)
Re: Conferences, Part 1 of 2
I don't believe they're programmer virtues, either; merely often just side-effects of the "classic" programmers.

The theses here vary wildly. I've seen some be as little as 60 pages. I also saw one of my friends write a 120-page thesis on a problem that had been an open computational geometry problem for thirty years. But I'd imagine that's true of a lot of places.

Here, at least, the first year is considered the most work. Theses generally aren't quite the equivalent of five courses. It's just that, like in most things, there's always a little more that could be done, you know?

Right now, my basic area is web services and the semantic web. However, I've got a number of elements of applied AI in my work. Right now, my work centres on clients reasoning over the past experience of themselves and other clients to select among syntactically identical web services, with the reasoning being done using naive Bayes. The past experiences are written up in RDF and OWL, taking advantage of W3C recommendations for the semantic web, as web services are little more than web-accessible applications (and thus, it seemed appropriate).

So basically, I'm somewhere between applied AI, systems, and networking applications. I've had some experience with formal AI as an undergrad (phase transitions and the like), and I'm a part of the summer GMG (graphical models group) reading group. We meet weekly for a couple of hours to read papers on Bayesian networks, Bayesian learning, and things related that. This week was a paper by MacKay on Monte Carlo methods.

(Sorry if this and my other post are a bit short - I'm technically on vacation right now [visiting my girlfriend in Manitoba], and her computer is none too good.)
banazir
Jul. 8th, 2004 05:54 pm (UTC)
Conferences, Part 2 of 2
Get a paper into IJCAI-2005 or a satellite workshop thereof, maybe?
Might be possible, but unlikely. IJCAI is mostly more formal AI, isn't it? My stuff is pretty applied - modelling, service selection, classification, etc.
Well, there are machine learning, negotiation, agents, and similar tracks of IJCAI. Also, in the USA there is IAAI (Innovative Applications of AI), which is co-located with AAAI.

Is that what the workshops are for? (I don't know much about the whole conference process - I submitted my first paper about a month ago, and don't get word until July 23)
Well, those, too.
Take a look at AAAI and see the Edmonton-2002 site for links to other conferences (UAI, KDD). In our area, there's also ICML, NIPS, and GECCO among the high-quality conferences.

Technically, I have some unfinished research in a completely unrelated area that I could submit (on Bayesian approaches to phase transitions in NP-hard problems) that was my undergrad thesis, but it's very unpolished, and probably needs way more experimentation than we did.
Interesting topic.
Have you read the Kautz, Horvitz, et al. paper from IJCAI 2001?

Want an auto-reminder?
Nah, that's okay. The cool thing about grad studies here is that registration has a far greater window than undergraduate studies. When I was an undergrad, if you weren't on the phone registration system the moment you were eligible, most of the classes would be filled. If the class I want to take is filled, I can just go down the hallway and get Dr. Vassileva's signature. :)
Heh, we have pretty open registration (especially for elective courses), except that instructor permission is required for grad courses.

--
Banazir
prolog
Jul. 9th, 2004 02:41 pm (UTC)
Re: Conferences, Part 2 of 2
I'll take a look at those. I'm not sure if I've read the Kauvitz, et al. paper. I don't think so, but it's possible. It's not in my dog-eared stack, anyway.
banazir
Jul. 8th, 2004 04:48 pm (UTC)
Thank you
I think it's lovely that you care.
I don't think it's weird at all that you feel a little nostalgic and little sad.

:o)
This kind of mass graduation only happens every once in a few years.
I had one in December, 2000, and one in June, 2003.

These are people you've taught and helped and encouraged and nurtured intellectually. It's like little birds leaving the nest. Of course you're going to be sad. And, sorry to sound so cheesy, I think it's great you care about them.
That's very astute. I think the analogy is fairly apt, particularly considering the way I think of my students.

I knew there was more than one reason I liked you, Banazir.
Aww...
Thanks, M.

--
Banazir
myng_rabbyt
Jul. 9th, 2004 05:57 am (UTC)
Re: Thank you
:o)
This kind of mass graduation only happens every once in a few years.
I had one in December, 2000, and one in June, 2003.


At least you don't have your heart twisted out every year. I think it must be hard for those folks who graduate students *every year*. My mom got over it after a while. But then, she only had most of her students for a year, and those little bastards were ungrateful. :P

I hope, when I finally reach that point of teaching on the college level (regular classes, not the mass instructional sessions I currently do), that I'll care enough to feel pride and pain when my students fly the nest.

That's very astute. I think the analogy is fairly apt, particularly considering the way I think of my students.

I'm surprisingly very adept at language. I just appear flakey, to throw people off. ;)

I knew there was more than one reason I liked you, Banazir.
Aww...
Thanks, M.


You're welcome, B. =)
( 22 comments — Leave a comment )

Latest Month

December 2008
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

KSU Genetic and Evolutionary Computation (GEC) Lab

Teunciness

Breakfast

Science, Technology, Engineering, Math (STEM) Communities

Fresh Pages

Tags

Page Summary

Powered by LiveJournal.com
Designed by Naoto Kishi