calebdame on Nostr: Data scientist here. I can assure you that llama are not replacing many existing ML ...
Data scientist here. I can assure you that llama are not replacing many existing ML applications, mostly new ones. Tree based models still dominate with tabular data structures (I worked in credit fraud aand we just used boosted tree algos). If llms are used in these applications to generate insights it’s usually hallucinations. If useful at all in these applications it’s them being used as orchestrators to train purpose-build classifiers regressors or networks.
After 3 years of R&D with all types of network structures and alternative algos, nothing beat a boosted tree in our client’s credit card fraud data sets
Published at
2025-01-25 06:16:54Event JSON
{
"id": "56a6384a8a2d336f79f4cfb70710076e7d47f709ebeb6469a61bc0b7cf9a1d4f",
"pubkey": "ec2e0ad4f2d20930b8a4895462016d2d2fd62c7c60523b7d8ec188636bca24c9",
"created_at": 1737785814,
"kind": 1,
"tags": [
[
"e",
"7c9dbee01d6cf870c77705e7a8da9e93635aad9b5f665f8293fe5f82380853ed",
"wss://nostr-pub.wellorder.net",
"root"
],
[
"p",
"50d94fc2d8580c682b071a542f8b1e31a200b0508bab95a33bef0855df281d63"
]
],
"content": "Data scientist here. I can assure you that llama are not replacing many existing ML applications, mostly new ones. Tree based models still dominate with tabular data structures (I worked in credit fraud aand we just used boosted tree algos). If llms are used in these applications to generate insights it’s usually hallucinations. If useful at all in these applications it’s them being used as orchestrators to train purpose-build classifiers regressors or networks. \n\nAfter 3 years of R\u0026D with all types of network structures and alternative algos, nothing beat a boosted tree in our client’s credit card fraud data sets",
"sig": "052c7de3242acb1b3ef2b3da8936adb1c691b7a23b9ca59bca5c6e81ee14af4e28500d5ba5344b17996fb755fe4577ad9d189005eeebaec542edc27abf53e8ea"
}
