Gearing up for start of term

[R. Paul Wiegand]

Greetings,

I hope everyone had a productive and/or fun Summer.

I am wondering who remains interested in having regular Algorithms &
Theory group meetings this year, and whether there are preferences
regarding when we meet? Early information is good so that I can get
in a request for a classroom for the Term, etc.

Additionally, I'd like to ask people to think about potential topics
that they'd like to discuss or see discussed, etc.

Thanks,
R. Paul Wiegand.

[Chen-Fu Chiang]

Dr. Wiegand, 

Thanks for the information. I am interested and I will participate more this term. 

As for topics, I think since last semester people already learned about random walk and some probability theory, maybe it would good to see its applications in various fields, such NLP, Bioinformatic, Quantum and Statistics. 

Of course, people can also do something new. For instance, implementing those algorithms, share the code and run a demo. After all, algorithms need to be implemented and being theoretical is as important as being practical. 

Chen-Fu Chiang 

[Abdullah Mahmud]

I want to participate every meetings this fall, but this summer went very unproductive for me with no study and research activities. So I do not have any concrete idea about what should I present. May be I will find something to present in the middle of Fall.

 

-satej

On Mon, Aug 8, 2011 at 9:35 AM, R. Paul Wiegand <rpwi...@gmail.com> wrote:

[Stephen Fulwider]

I'm still interested in meeting as well. I'm happy to meet anytime except MWF 12:30-2:30.

I'm in a similar situation to satej in that I haven't given any thought this summer to research, but am looking forward to getting back into it this semester. Maybe we could meet on the 2nd or 3rd Friday of the semester as a kick off meeting? I'm open to other suggestions as well.

Hope everyone had a great summer and look forward to seeing everyone soon!

Stephen

[Nadeem Mohsin]

I'd be interested in participating this term as well. I'm going to be spending some time this semester boning up on probabilistic graphical models (which are getting pretty big in NLP), so that's a pretty cool bit of machine learning that we can discuss.

Nadeem.

On Mon, Aug 8, 2011 at 9:35 AM, R. Paul Wiegand <rpwi...@gmail.com> wrote:

[Abdullah Al Mahmud]

Great. I think in our first session Nadeem can start lecturing on probabilistic graphical models? I have not explored that much but I would be pretty much interested to learn from you.

-satej

[Nadeem Mohsin]

Well, perhaps not this week, but I'm certainly up for next week. :) It's probably best to wait a week or so, to make sure everyone's schedules are properly nailed down. We should probably discuss topics and so on this week.

Nadeem.

[R. Paul Wiegand]

Greetings all,

Several of the ideas discussed in this thread seem plausible and
interesting for us to pursue. I think we should remain flexible and
let those who want to explore a take responsibility for it. My main
objective is to try to maintain a "space" where students interested in
theory can get together and explore theory-oriented topics in a more
interactive and self-paced / self-directed way than they might
coursework. The group itself should decide what it wants to do.

These things can be among the conversations on the first meeting day,
of course. worth (or so) of a topic on which to lecture?

As to when we start to meet, I agree that this first week seems
unlikely. Moreover, I'll be out of town next week. So why don't we
begin in the third week of term? Nadeem: If we meet first on the
third week of term, might you have an hour's worth (or so) of material
to discuss?

But we still need to settle on a day before I can request a room. Any
objections to meeting on Thursdays from 2:30p -- 4p?

I arrive at this suggestion by considering the following constraints:
* MWF, 12:30p -- 2:30p is unavailable;
* F, 3:30p -- end of day is unavailable;
* W, 3p -- end of day is not preferable;
* TR, 12p -- 1:15p is unavailable;
* EITHER T or R, 2p -- 3:30p is unavailable;
* More generally, the position in the past has been that a morning
meeting is not conducive to turnout from graduate students.

Are there any other constraints I should consider?


R. Paul Wiegand.

[Nadeem Mohsin]

I'll probably have something by the third week of term. That should work.

Nadeem

[Abdullah Al Mahmud]

So what can be the final schedule? I think we are meeting from next week.

-satej

[R. Paul Wiegand]

I apologize for the delay; I was out of town with limited access to
internet.

I submitted a request for a classroom on Thursdays from 2:30p-4p last
week, but have not yet heard back about it. Hopefully we will have an
answer today; however, I'm well aware that we're running short on time
to "advertise" the meeting. If I get favorable word back today, I'll
send out an announcement today. If not, I'll post back to the group
indicating that we'll hold off a week until the meeting location is
settled.

R. Paul Wiegand.

[R. Paul Wiegand]

I have not yet heard back from the people with whom classrooms are
reserved, so I'll have to follow up with them today; however, I think
it is too late at this point to proceed today. Let's reschedule for
next week, if Nadeem is amenable. That way, we have time to resolve
whatever the hangup with the room reservation is, and we have time to
publicize the meeting.

I apologize for the problems we are having.

R. Paul Wiegand.

[Nadeem Mohsin]

Works for me.

Nadeem

[R. Paul Wiegand]

Greetings,

We weren't able to get a classroom, but we got HEC 356 for the
semester on Thursdays, 2:30p-4p, except Sept. 29, where we have HEC
450 reserved. I'll send out a notice about this Thursdays meeting to
the grad list (and here) later today. Nadeem: Can you send me a
sentence or two about what you plan to talk about?

Thanks,
R. Paul Wiegand.

[Nadeem Mohsin]

I was thinking of something along the lines of some basic background on Bayesian inference and joint distributions, from which I can build up to the basics of graphical models. I should be able to talk a bit about the different kinds of models, how we use them, and maybe mention a few applications.

Nadeem.