connor @notwa@cybre.space

oh god i forgot that i wrote this

my code is bugged but the output is pretty

i wrote a python version of big_something and it's *so* much shorter. it should be easier to reason about.

ok now the results make a lot more sense. i guess this is division in the finite field

i still don't know why this function does what it does but i gave it a fancy state visualization

there it's finally done. when one of the numbers in the product key equals the hash of the whole(?) key, the key is good: the two at the bottom here are equal.

oh huh venn diagrams are actually not a bad way of visualizing 3-in 1-out truth tables, here's (a&b)|((a|b)&c)

epiphany

rolled up. no idea what this is doing.

tempted to reverse engineer this old thing

stupid in-joke ignore this

i've been busy working on this bus blueprint, still only like half done

linearly inconvergent

???

rotated rastrigin contours

OOOH you have to click on the thing on the sidebar, i thought it was just like an anchor link for something on the same page :/

this simple thing made me laugh quite a bit

oh no what happened here

ok, now this is weird. an evolved algorithm decides to use the *mean* of a network's weights as its learning rate.

via: https://arxiv.org/abs/2003.03384

found a bug in this random connectivity algo where it tries to connect the same node twice, but it otherwise seems to work well. the bug is increasingly rare as node count increases and i have a couple hundred on each side in practice so it's not too big of a deal at this moment.