|
|
Post by Lars Ericson on Aug 10, 2015 5:55:52 GMT -8
Just curious: 1. No statue? 2. 10,000 XXX prize from years past, no longer. We could at least pool for a ₫10,000 prize mounted in a framed certificate? I would be happy to chip in.  3. No writeup of the winner? Gameplayer design, bio of the author? 4. Alternate prize suggestion, even better, something along these lines:  |
|
|
|
Post by Steve Draper on Aug 10, 2015 9:31:26 GMT -8
There's a cup, which I will be transferring to Richard the week after next, as I'm meeting him for dinner while I'm on vacation near where he lives.
I agree that it would be nice to see more (technical) write-ups about the players, but that's really up to the authors, and not the competition organizers.
|
|
|
|
Post by Lars Ericson on Aug 10, 2015 11:08:34 GMT -8
This is the most introverted, shy sporting event I've ever experienced.
|
|
|
|
Post by maciej on Aug 11, 2015 8:04:51 GMT -8
Please correct me if I am wrong, but in the first two years (2005 and 2006) where the 10000$ prize was around, the winner was required to publish a paper about their approach in order to receive the prize.
I think that was the initial idea, to share what the state-of-the-art is.
I have dropped out of the competition, because of numerous of reasons, primarily because I hate using propnets in GGP and they have become the absolute necessity and the single most important factor for doing well. The selection of games seem to be propnet friendly.
The other main reason being lack of time and I wasn't sure whether the competition would be in the summer (until it was announced) since the AAAI conference moved to a winter schedule.
|
|
|
|
Post by Lars Ericson on Aug 20, 2015 17:15:18 GMT -8
(a) You don't have to use propnets. I'm pursuing a Clingo-based solution and I don't think I'm wasting my time. (b) I have offered to put up a 10,000 Dong prize to set things right on the prize end, at least unit-wise. (c) Galvanize did put up a description of the player on its (public) repository, so that part's right. (d) This competition appears to have gone from a Stanford-centric enterprise to a GDL diaspora centric one. It seems a bit more ad hoc. Pages don't get updated much. The games.stanford.edu site doesn't even have an announcement that the competition is over and the winner is Galvanize, though you can sort of guess that if you were intimately involved with the competition. As a diaspora-centric activity, there is not, I would guess, the entree to prestigious journals that would come with a Stanford professor's guidance and endorsement. So instead of journal articles and some academic momentum, we have engineer/enthusiasts here and there and a Github entry. But that's OK! |
|
rxe
Junior Member
Posts: 61
|
Post by rxe on Aug 21, 2015 9:44:30 GMT -8
Lars, I believe you have confused the players Generalize and Turbocho.
|
|
|
|
Post by Lars Ericson on Aug 21, 2015 11:38:22 GMT -8
Sorry...Galvanize?
|
|
|
|
Post by gscottj on Oct 15, 2015 9:25:53 GMT -8
[...] (d) [...] As a diaspora-centric activity, there is not, I would guess, the entree to prestigious journals that would come with a Stanford professor's guidance and endorsement. So instead of journal articles and some academic momentum, we have engineer/enthusiasts here and there and a Github entry. But that's OK! It's OK, but it's not necessarily great. I started into GGP from an academic point of view at the beginning of the year, and only just stumbled onto this forum today. Papers and conferences are how academics advertise about things like this, and if resources like this don't wind up being known about, then worse than a diaspora, there might be a fracturing of the community. I think we all lose if we aren't all together, keeping each other on our toes. |
|
|
|
Post by Lars Ericson on Oct 27, 2015 20:12:25 GMT -8
The issue is that this was apparently a big deal with a $10,000 prize a while ago. Stanford/Coursera has a GGP course which this forum is closely connected to. Students taking the Coursera course are not much connected with Stanford or with academia. The TAs in the course are from the Coursera population, not Stanford or academia per se. One of the chief GGPBase architects is Sam Schrieber who is at Google and hence more industry than academic. When you take the course on Coursera you are more likely to hear from a TA or Sam Schrieber than from the professor, now that the course has gone through several iterations and perhaps the professor's focus has moved on from Coursera.
Having been in the academic loop once upon a time (I was in CMU CS Dept undergrad and have a PhD in CS from NYU), I have direct experience with Stanford, the AAAI conference, and the sense of star power and wattage that comes from being on campus in a place like that. Coursera is a generous enterprise, but it doesn't put you on campus, doesn't make you a grad student, and doesn't give you that professor as a thesis advisor. Professors when they are working on their own account, and advancing the careers of their students to advance their own career, put a lot more effort into promotion than they would for activities conducted by non-academics off of their radar. It's just not the same.
So once the professor gets over the thrill of having a MOOC with a ton of students, and does it a few times, then perhaps that MOOC is no longer part of his primary focus. That means the $10,000 disappears and, in a AAAI off-year, that means that the competitive demonstration focus disappears, or at least moves away from the Coursera course.
The end result is that GGPBase and GGP.org become folklore among ex-students of the Coursera GGP course, the competition becomes something run by aficionados on a shoestring, there is a unique trophy that gets passed around, and the victory lap is a dinner in Seattle or wherever the competition organizer is if you happen to live in that town. Which is all good. It's not the same as being a current grad student in a top 3 CS Dept and living on or near campus and having a fellowship and a thesis advisor on campus.
So rather than lament the passing of gestation and infancy for GGP, we should be happy that it has survived and can take care of itself and live on, even if it is a homeless child living in the streets of Austin, Texas.
|
|
