Re:Generative AI is overrated, long live old-school AI
by jacksnipe on Wednesday March 15, 2023 @02:35PM
(#35172722)
We loved them in medical testing. Very explainable models.
Re:Generative AI is overrated, long live old-school AI
by teruakohatu on Wednesday March 15, 2023 @03:32PM
(#35173550)
> do people actually use SVMs anymore?
Yes they are. They allow for non-linear decision boundaries and more dimensions than rows of data, which for many other ML methods is a problem.
Linear regression, logistic regression, SVM and CART decision trees are all still very popular in the real world where data is hard to come by.
Re:Generative AI is overrated, long live old-school AI
by superdisk on Wednesday March 15, 2023 @04:19PM
(#35174177)
> but the first and last time I heard of SVMs was in college, by professors who were weirdly dismissive of these newfangled deep neural networks, and enamored by the "kernel trick."
LOL. Exact same experience in my college courses. Glad to know it's universal.
like, regression, sure - because it's a tool to measure how well a hypothesis (polynomial function) matches the data (points.) and CNNs are still foundational in computer vision. but the first and last time I heard of SVMs was in college, by professors who were weirdly dismissive of these newfangled deep neural networks, and enamored by the "kernel trick."
but aren't SVMs basically souped up regression models? are they used in anything ML-esque, i.e. besides validating a hypothesis about the behavior of a system?