This is a reference to the "wanna feel old?"-style comics, another recurring xkcd theme, for example [0], [1]. But yes, feeling old sucks.

[0] https://xkcd.com/1393/

[1] https://xkcd.com/1686/

Is there a diagnosis for the tendency to collect diagnostic labels and use them to absolve oneself of responsibility?

On further thought, not sure if this would solve the problem or exacerbate it...

Hofstadter introduces several sequences in GEB, [0] may be an interesting submission on its own but I was especially captivated by this self-referencing one. Plus a title including both Hofstadter's sequence and a description is too long for HN and I preferred the descriptive one

[0] https://en.wikipedia.org/wiki/Hofstadter_sequence

I meant the title as it appears in OEIS, not as it appears on HN :)

Ah :)

From my (limited) experience the OEIS titles lean strongly to the descriptive side too. But maybe also to avoid ambiguity regarding to which one is it from his sequences?

I too think they're gorgeous. They're by Mark Ferrari, you can find more here[0].

[0] http://www.effectgames.com/demos/canvascycle/

8 Bit & '8 Bitish' Graphics-Outside the Box

https://www.youtube.com/watch?v=aMcJ1Jvtef0

> In this GDC 2016 talk, Terrible Toybox's Mark Ferrari discusses and demonstrate some of his techniques for drawing 8 bit game graphics, including his celebrated methods for use of color cycling and pallet shifting to create complex and realistic background animation effects without frame-animation

The lost art of color cycling - Animating with color

https://www.youtube.com/watch?v=LUbrzg21X9c

> Color cycling was born out of hardware restrictions but out of it came some beautiful artworks that characterized the 80s and 90s. In this video we will learn more about how this effect was produced and experience first hand the beautiful artwork created with it.

amazing, i love interesting rabbit-holes into the past initiated by engaging with HN :D immediately put them on watch later.

oah, thanks a lot! love his artstyle, gotta look him up :)

not sure what about his animations captures my eyes, but i think he makes great use of 2D depth perception and color palettes.

Really neat visualization! And thanks for the tip on MCC.

Out of curiosity I plugged it to the same visualization (performance vs. class weight when optimized with BCE) and it behaves similar to F1, i.e. best without weighting.

Yeah it gets tricky. I think eventually it has to be about tradeoffs - no ML system can be 100% correct. I do think there's a "right" decision (up to a point) in the context of the product or business.

A fully connected layer has different weights for each feature (or position in input in your formulation). So the word "hello" would be treated completely differently if it were to appear in position 15 vs. 16, for example.

Attention, by contrast, would treat those two occurrences similarly, with the only difference depending on positional encoding - so you can learn generalized patterns more easily.

I think that this is the explanation I needed, thanks!

If the questions were pre-determined, which they're usually not. Reminds me of Huffman coding and the reason that compression challenges measure submissions looking at artifacts required to run them in addition to compressed size. I tend to agree with OP that this doesn't pass the smell test

Thanks!