[Cs5500] Fwd: [Cs4800] Number of games per avatar differs

Fri Nov 11 18:14:14 EST 2011

I second Haoran's suggestion, it is sensible.
The other avatar does not gain points because it did not use any of its
intelligence to kick the avatar out.

But we may have to check one condition - the avatars that is kicked out
does not come back and ruin the playground by passing null values or fair
values. I say this because some of these "kicking" conditions are checked
as part of the catch block when exceptions are generated.

On Fri, Nov 11, 2011 at 4:47 PM, Karl Lieberherr <lieber at ccs.neu.edu> wrote:

> Hi Haoran:
>
> you mean: when an avatar is kicked in a binary game, we let the other
> avatar win and the violating avatar lose independent of the reputation they
> have accumulated in this binary game.
> We keep the avatar in the tournament so she can play more binary games.
>
> An this involves only changing one line! That is good news. Let's make
> this change, test it and use it for the tournaments.
>
> This will make the undergraduates happy, I believe.
>
> -- Karl
>
> On Fri, Nov 11, 2011 at 4:12 PM, Haoran Wu <haoran at ccs.neu.edu> wrote:
>
>> Hi Prof.Karl,
>>
>> As my understanding, when an avatar is kick out in binary game, by this
>> time can we not update this avatar's opponent's score?
>>
>> Do you think it's fair to do this? If so, just a line's code change.
>>
>> Haoran
>>
>>
>> ----- Original Message -----
>> From: "Karl Lieberherr" <lieber at ccs.neu.edu>
>> To: "Managing Software Development" <cs5500 at lists.ccs.neu.edu>
>> Sent: Friday, November 11, 2011 11:09:03 AM GMT -05:00 US/Canada Eastern
>> Subject: [Cs5500] Fwd: [Cs4800] Number of games per avatar differs
>>
>>
>> It looks like that the undergraduates will insist that our current
>> kicking strategy is unfair.
>>
>>
>> How much effort is it to change to the "kick from binary game only"
>> policy?
>>
>>
>> -- Karl
>>
>>
>> ---------- Forwarded message ----------
>> From: Greg I Kerr < kerr.g at husky.neu.edu >
>> Date: Fri, Nov 11, 2011 at 10:39 AM
>> Subject: Re: [Cs4800] Number of games per avatar differs
>> To: Karl Lieberherr < lieber at ccs.neu.edu >
>>
>>
>>
>> Hi,
>>
>>
>>
>> I am still not convinced that a tournament with the current rules is
>> fair. Why should an avatar that can successfully defend, refute and
>> strengthen all claim miss out on points because someone else was ejected?
>> It is worth noting that almost universally in sports, if your opponent is
>> ejected from a match, you are automatically given a Win.
>>
>>
>> Since a round-robin tournament dictates that all teams play every other
>> team, if we kick people out it is hard to call this a round-robin
>> tournament.
>>
>>
>> Regards,
>>
>>
>> Greg
>>
>>
>>
>>
>> On Fri, Nov 11, 2011 at 10:06 AM, Karl Lieberherr < lieber at ccs.neu.edu >
>> wrote:
>>
>>
>> We make an excursion from algorithm design to game mechanism design.
>>
>>
>> Dennis has a convincing argument for kicking avatars only out of a binary
>> game and not out of the tournament.
>> We have been too harsh with the avatars!
>>
>>
>> But we can run fair tournaments with the current rule, although it is not
>> perfect as Dennis and David point out.
>> We have to investigate how much effort it is to change the rule to
>> Dennis' rule.
>>
>>
>> -- Karl
>>
>>
>>
>>
>> On Fri, Nov 11, 2011 at 9:42 AM, Dennis Zografos < dzog at ccs.neu.edu >
>> wrote:
>>
>>
>> Regardless of whether or not that is true (it's not, by the way), it
>> seems obvious
>> to me that, in a full round-robin style tournament, no player should EVER
>> be ejected
>> from the tournament entirely.
>>
>> If Avatar X consistently exhibits "really bad" scientific behavior, it
>> will be kicked
>> out of (lose) every individual game it plays against valid avatars. All
>> the other
>> competitors have the opportunity to gain win points against Avatar X.
>>
>> Ejecting players from the tournament destroys the sanctity of the R-R
>> scoring system.
>>
>> (unless you want to start redacting points won against an ejected avatar,
>> but that
>> seems rather convoluted in this context)
>>
>> (alternatively, if you *really want* to eject players, wins should be
>> auto-awarded to
>> all the players who didn't get the "chance" to face them -- if there's a
>> good reason
>> for this in an HSR context, I'd like to hear it)
>>
>> -Dennis
>>
>>
>>
>>
>> ----- "Madhuvanthi Balasubramanian" < balasubramanian.m at husky.neu.edu >
>> wrote:
>>
>> > Matthew and David,
>> >
>> >
>> > When an avatar is kicked out, it is kicked out forever, from the
>> > tournament.
>> > I don't think there's any case where an avatar is kicked out from a
>> > round and then comes back to play in another round.
>> >
>> >
>> > On Fri, Nov 11, 2011 at 9:18 AM, Matthew Strax-Haber <
>> > straxhaber at gmail.com > wrote:
>> >
>> >
>> >
>> > Prof. Lieberherr,
>> >
>> >
>> > I think David was referring to the fact that in some cases an avatar
>> > who performs 'bad scientific behavior' is kicked out from a round, and
>> > in others from the whole tournament.
>> >
>> >
>> >
>> >
>> >
>> > Best regards,
>> > --
>> > ~ Matthew Strax-Haber
>> > Northeastern University
>> >
>> >
>> >
>> >
>> >
>> > On Nov 11, 2011, at 5:05 AM, Karl Lieberherr wrote:
>> >
>> >
>> > Hi David:
>> >
>> >
>> > There is no difference.
>> >
>> >
>> > There are two kinds of "punishment" behavior in SCG: for bad
>> > scientific behavior you lose reputation
>> > and for really bad scientific behavior, you get kicked out.
>> >
>> >
>> > Bad scientific behavior is, e.g., not defending your claims.
>> > Really bad scientific behavior is, e.g., to provide a solution that is
>> > NOT valid for the given instance. The valid predicate is defined in
>> > the playground definition.
>> >
>> >
>> > Thank you for all your great emails.
>> >
>> >
>> > -- Karl
>> >
>> >
>> > On Thu, Nov 10, 2011 at 11:30 PM, David Richards < dirich at ccs.neu.edu
>> > > wrote:
>> >
>> >
>> > What are the different criteria for getting kicked out of a single
>> > game versus getting
>> > kicked out of the tournament altogether?
>> >
>> > Thanks,
>> > -David
>> >
>> > On Nov 10, 2011, at 11:17 PM, Madhuvanthi Balasubramanian wrote:
>> >
>> > > David,
>> > > The avatar that was kicked out was iGalaxy. iGalaxy had a chance to
>> > play against dirich, but not 5150, hence the difference in # of
>> > tournaments.
>> >
>> >
>> >
>> > >
>> > > On Thu, Nov 10, 2011 at 11:10 PM, Madhuvanthi Balasubramanian <
>> > balasubramanian.m at husky.neu.edu > wrote:
>> > > David,
>> > > It is possible that some avatar may have been kicked out in the
>> > midst of the tournament.
>> > > As a result, 5150 may not have been able to compete with that
>> > avatar, while you may have.
>> > > This is a possibility.
>> > > I still have to look through the files to see who was kicked out and
>> > when.
>> > >
>> > >
>> > > On Thu, Nov 10, 2011 at 11:02 PM, David Richards <
>> > dirich at ccs.neu.edu > wrote:
>> > > Why is it that the number of games differs per avatar? For example,
>> > in the last
>> > > tournament my dirich avatar played 21 games winning or tying all of
>> > them for
>> > > a total of 63 points. However, 5150 only got to play 20 games,
>> > winning or
>> > > tying all of them for a total of only 60 points.
>> > >
>> > > Shouldn't each avatar play the same number of games?
>> > >
>> > > -David Richards
>> > > _______________________________________________
>> > > Cs4800 mailing list
>> > > Cs4800 at lists.ccs.neu.edu
>> > > https://lists.ccs.neu.edu/bin/listinfo/cs4800
>> > >
>> > >
>> > >
>> > > --
>> > > - Madhu Murali
>> > >
>> > >
>> > >
>> > > --
>> > > - Madhu Murali
>> > >
>> > >
>> > >
>> > > --
>> > > Thanks
>> > > - Madhu
>> > >
>> > >
>> > >
>> > >
>> > > --
>> > > Thanks
>> > > - Madhu
>> >
>> >
>> >
>> > >
>> >
>> >
>> > _______________________________________________
>> > Cs4800 mailing list
>> > Cs4800 at lists.ccs.neu.edu
>> > https://lists.ccs.neu.edu/bin/listinfo/cs4800
>> >
>> > _______________________________________________
>> > Cs4800 mailing list
>> > Cs4800 at lists.ccs.neu.edu
>> > https://lists.ccs.neu.edu/bin/listinfo/cs4800
>> >
>> >
>> >
>> >
>> >
>> > --
>> > Thanks
>> > - Madhu
>> >
>> > _______________________________________________
>> > Cs4800 mailing list
>> > Cs4800 at lists.ccs.neu.edu
>> > https://lists.ccs.neu.edu/bin/listinfo/cs4800
>>
>>
>> _______________________________________________
>> Cs4800 mailing list
>> Cs4800 at lists.ccs.neu.edu
>> https://lists.ccs.neu.edu/bin/listinfo/cs4800
>>
>>
>>
>>
>> _______________________________________________
>> Cs5500 mailing list
>> Cs5500 at lists.ccs.neu.edu
>> https://lists.ccs.neu.edu/bin/listinfo/cs5500
>>
>
>
> _______________________________________________
> Cs5500 mailing list
> Cs5500 at lists.ccs.neu.edu
> https://lists.ccs.neu.edu/bin/listinfo/cs5500
>
>

-- 
 Thanks
- Madhu
-------------- next part --------------
HTML attachment scrubbed and removed