Hacker News Reader

Comment by novok | Hacker News Reader

Dimension 1 and dimension 2. They reduced a multidimensional vector output to 2 dimensions to make it visualizable.

https://en.wikipedia.org/wiki/T-distributed_stochastic_neigh...

IMO this study feels like it's saying "facial recognition AI model outputs vectors where female and male faces cluster closer to each other than not". Which to a typical human is stupid obvious.

The author is spinning up drama to get more engagement in his article IMO.

View on HN

> This study feels like it's saying "facial recognition AI model outputs vectors where female and male faces cluster closer to each other than not".

In this analogy, it would be like concluding that "androgyny does not exist."

I'm curious how they handled some of the complicated selection and coding issues around trans and intersex people. Were those groups excluded, or directed in responding in one way or the other? The study is large enough to expect a few of these cases.

The ability to perfectly separate the groups makes me suspicious of a data leakage or methodological error. Perfect accuracy rarely is.

It might not say much about how male and female brains are different, but just being able to measure whether a brain has male or female brain activity could be a powerful health tool. For example, what if we could use this to determine ahead of time who will respond positively to gender-affirming care and who will respond negatively? It could save lives.

Reply on news.ycombinator.com