npub15s…e8xgu on Nostr: Lots. Been infatuated with deep learning for ~10 years. RL is unfortunately ...
Lots. Been infatuated with deep learning for ~10 years. RL is unfortunately data-inefficient, but so far the best algorithm we have for some things. Its data inefficiency isn't really endemic though, it's just that we don't yet have scalable mechanisms for learning abstractions, only distributions.
I'll copy my last slime mold-related note:
Published at
2024-12-28 08:15:57Event JSON
{
"id": "fb76dc345c7405bd1ca2cfa56063cf356f68cc9227467d54cc4bd2bc0a59f63c",
"pubkey": "a43b0118fd72492f2ba11290cccb27418b1fdbb7ce3a122d229404e57a75975a",
"created_at": 1735373757,
"kind": 1,
"tags": [
[
"e",
"904a0315e4aa9fb6ee7a3155253c3a578d781168fbc62b4e69aac2484b1bcf87",
"",
"root"
],
[
"e",
"1efd1cca012a96f6ab24cc0a245dc62527412c3c21e9d45da8d26a0d7417e9a3"
],
[
"e",
"a635ed1b1cc79c48d341a0e5b2c99d4c9fac78d34d01e2351cd08973f71b00a0",
"",
"reply"
],
[
"p",
"0a69cf2560597cd4dfff9a75f40261d902a91b139cdacea10d54a52b43219250"
],
[
"p",
"a43b0118fd72492f2ba11290cccb27418b1fdbb7ce3a122d229404e57a75975a"
]
],
"content": "Lots. Been infatuated with deep learning for ~10 years. RL is unfortunately data-inefficient, but so far the best algorithm we have for some things. Its data inefficiency isn't really endemic though, it's just that we don't yet have scalable mechanisms for learning abstractions, only distributions.\n\nI'll copy my last slime mold-related note:",
"sig": "e4f6463e7f63a7167d21370c91bfd4449a2843160fc115d1cff8f7e8d3f579af508aceb6526ae0f67597c5ea829a8c6029c01a239cb05f8a50f83d2b33051a60"
}