4 ms·
People routinely use lagrange polynomials, fourier approximations, etc. with 10^10 or more coefficients and many more data-points. That's often orders of magnit
by fluffything 8y ago
People routinely use lagrange polynomials, fourier approximations, etc. with 10^10 or more coefficients and many more data-points. That's often orders of magnitude larger than what training most neural nets for AI need.