4 ms·
Kronecker in (O(p(n)^3): A Novel approach to 80-year-old P-hard problem [pdf]
- bon-cdp 1y agoHad this idea while working on wreath-sheaf attention to use categorical bootstrap to approach Kronecker Coefficient computation. Code is uploaded to https://github.com/bon-cdp/heir/tree/experimental/unified-sheaf-attention https://github.com/bon-cdp/heir/tree/experimental/unified-sh... if anyone is interested.
- fn-mote 1y agoBlockchain proponent [1] with no visible github activity posts a paper full of garbage - no proofs, barely any definitions. No methods. Nobody could believe this. [1]: https://sf-labs.co/ https://sf-labs.co/
- bon-cdp 1y agoOverall you are so right to be skeptical. This is something I'm actively working on and just wanted to share the idea as I validate (sorry if that's wrong) The crypto thing (trustdog) was just a hackathon thing i recently did (which originally was supposed to be a stripe product but they restrict escrow) and i work on other stuff a lot so i dont always have time to update the personal site. its more of an idea with some experiment (which you can see and run yourself!) and not a "paper". I made my first contribution to google/HEIR that got pulled the other week, but there is a lot of self activity because I do this for fun! im a mechanical engineer by profession. I dont mind if no one believes it. Just wanted to share something I thought was cool. Didn't mean to offend anyone in any way.
- bon-cdp 1y agoA little update after thinking about it more, since its dependant on the character table for the cyclic subgroup with k (O(k)) (this is kinda the whole idea - to do so with Frobenius Reciprocity); o(p(n)^3) is kinda inaccurate, as it ends up being that character group operation plus around O(k*p(n)) plus potentially another p(n).