Dear
Harnidh, Yosem-
In the close of last chapter of Algorithms to Live By
<http://algorithmstoliveby.com/> about the challenges and potentials of
Game Theoretic Mechanism Design by way of the Revolution Principal
<http://faculty.haas.berkeley.edu/rjmorgan/phdba279b/Class%201%20Introduction.pdf>
,
Brian Christian <http://brianchristian.org/> and Tom Griffiths
<http://cocosci.berkeley.edu/tom/index.php> mention the concept of
Computational
Kindness <https://egtheory.wordpress.com/2016/06/30/computational-kindness/>.
The point I make here it that it is possible to construct an N-Way,
spoiler effect free vote tallying systems that is always "computationally
kind" to all voters of the electorate. That is, a voting system where
every voter's best strategy is always to express the truth about there
preference rankings as placed upon a cardinal personal utility scale..
furthermore such an automated strategic voting system can be made to
operate entirely within a information leveled context, where any issues of
strategic information asymmetry would be entirely mitigated.
While for some combinations of ballot input (when there is not a clear
cut pairwise head to head Condorcet win against all other contenders), as a
good understanding of Gibbard–Satterthwaite theorem
<https://en.wikipedia.org/wiki/Gibbard%E2%80%93Satterthwaite_theorem>,
suggests, some amount of departure from a fully deterministic outcome is
required to defeat any potential of strategic vote gaming, the introduction
of this fully hidden randomization variable can be configured in such a
fashion as to minimize the field of ultimate contention as well as the
degree of randomization ultimately required..
That is: if we define the election outcome to be at the sole
equilibrium of automated strategic probabilistic (fuzzy) approval voting
game, that minimizes this width of a linear "viability margin" we can solve
for the this unique probability that minimizes the amount of randomization
required to defeat any possibility of strategic vote gaming... and thus
provide the social choice solution that is as close as is theoretically
possible to a deterministic outcome.
Furthermore if one then combines the above concept with a voter
specified strategic moderation parameter as suggest in this paper
<http://www.votingmatters.org.uk/ISSUE27/I27P1.pdf>, a very attractive
voting system can be constructed.
Hope this helps..
all the best
-Peter
On Sun, Feb 4, 2018 at 6:09 AM, Yosem Companys <ycompa...@gmail.com> wrote:
> From: Harnidh Kaur <kharn...@gmail.com>
>
> I don’t think most of you know this, but I work in the development space
> and we’re always trying to find cool ways to make tech make the world
> better. I’ve been trying to read up more about the same. So, here.
>
> Is ‘kind technology’ a thing? Where people are trying to change gears of
> existing/incumbent technology to specifically serve vulnerable populations?
>
> Any existing ideas/something you’re working on/things you think SHOULD find
> a kind tech iteration. Gimme!
> --
> Regards
>
> Harnidh Kaur
> Lady Shri Ram College for Women '15
> St. Xavier's College, Mumbai '17
> Foreverawkwardandlearning.wordpress.com
> <http://foreverawkwardandlearning.wordpress.com/>
> +91-7718951383
>
>
> --
> Liberationtech is public & archives are searchable on Google. Violations
> of list guidelines will get you moderated: https://mailman.stanford.edu/
> mailman/listinfo/liberationtech. Unsubscribe, change to digest, or change
> password by emailing the moderator at zakwh...@stanford.edu.
>
--
Liberationtech is public & archives are searchable on Google. Violations of
list guidelines will get you moderated:
https://mailman.stanford.edu/mailman/listinfo/liberationtech. Unsubscribe,
change to digest, or change password by emailing the moderator at
zakwh...@stanford.edu.
