What if we embed genders in hyperdimensional vectors and train a neural network to classify someoneβs βuniqueβ gender
Hmm, interesting idea, but what if the kid incorporates a xenogender axis orthogonal to all previously known xenogender axes? My algorithm would project away this axis and it would have no effect on the computed output, but a neural network would have randomly determined weights connected to the input corresponding to that axis, resulting in unexpected results
what about being genderfluid? does the winner changes from time to time? does it counts as bigender? what if the fluidity includes non boy/girl genders?
In this case youβre just fucked lol.
If the kid has a relatively clear βaverageβ gender, (e.g. spends 50% of the time as simply female, 25% of the time simply xenogender, 25% simply demimale) and this average doesnβt change over time, you could compute the average gender and then use the aforementioned algorithm to solve. BUT if the kid tends to move genders over years or even decades youβre never gonna have a clear winner.
If the kid comes out as genderfluid your best bet is to just call it a draw and both players get their money back
Okay but do you at least agree you have to use multivectors? 1D vector gender is clearly insufficient
How would you even represent a gender as a 1D quantity? Even ignoring xenogender axes you need at least 2 dimensions
this implies gender does not have a z axis
Actually it doesnβt, it merely projects all genders onto the male/female plane. I donβt think this is incorrect for this purpose, for instance if we consider a xenogender entirely orthogonal to the two standard basis vectors, projection onto the male/female plane yields coordinates of (0,0). It would be reductive to say that this is equivalent to other genders that map to this point, however, it still correctly yields the result of a tie. This is the natural outcome of neither player betting on a xenogender
βgreater than" and βless thanβ are missing the βor equal toβ. A difference of exactly 1 should be sufficient for an absolute victory. Iβm also missing a definition of what negative numbers on an axis mean, since they are currently implied to reach absolute victories and are not excluded by the definition. Iβll show myself out π
We simply define the two dimensional space as an empirical observation, where the median man is located at position (0,1). So there potentially exist men at position (0.1,1.1) but the epicenter of masculinity is at this coordinate - similarly so for female. In the construction of this simplified geometric space we have no guarantees except the median property of the two most popular genders. Since nonbinary genders frequently defy notions of categorization there is no reason to believe that the projection of an extremely high dimensional gender into this space will yield numbers with any meaningful properties. All we can say for certain is that a Manhattan distance greater than or equal to one is sufficient for an absolute win.
And the moral of the story? Betting and gambling is haram.
