Thiago Carvalho on Nostr: 'We therefore set out to explore how low-complexity rules can give rise to networks ...
'We therefore set out to explore how low-complexity rules can give rise to networks that perform complex well-defined neural computations. Within the framework of artificial neural networks (ANNs), we seek to compress the complex connectivity (weight matrix) into a much smaller “genome.”'
https://www.pnas.org/doi/10.1073/pnas.2409160121Published at
2024-09-16 12:47:33Event JSON
{
"id": "c366bbb549e9013f681ad12a0471901b89a3910bdf23e92776862446b4ec41bd",
"pubkey": "376d961fc946cb96839a4d257809dc9e6d237e2d0cc158199b8cc5c5771939a6",
"created_at": 1726490853,
"kind": 1,
"tags": [
[
"proxy",
"https://qoto.org/users/cyrilpedia/statuses/113147304578849237",
"activitypub"
]
],
"content": "'We therefore set out to explore how low-complexity rules can give rise to networks that perform complex well-defined neural computations. Within the framework of artificial neural networks (ANNs), we seek to compress the complex connectivity (weight matrix) into a much smaller “genome.”'\n\nhttps://www.pnas.org/doi/10.1073/pnas.2409160121",
"sig": "d6fa1358ee486462473e0394041be88afe8550234127b553a8630a9bf069eab696772596f5f7742bd5844bb1f475a428a9ef8f6b0d2df3f69e8b4975c68f5b27"
}
