Polynomial Regression As an Alternative to Neural Nets

These guys are going after neural networks claiming that they are essentially equivalent to polynomial fitting. Thoughts? If no objections to that claim, I’ll go run some polynomial regressions.

“… one may choose to routinely use polynomial models instead of NNs, thus avoiding some major problems of the latter, such as having to set many tuning parameters and dealing with convergence issues. We present a number of empirical results; in each case, the accuracy of the polynomial approach matches or exceeds that of NN approaches. A many-featured, open-source software package, polyreg, is available.”

https://arxiv.org/abs/1806.06850

By Alexey Khazanov

 

Leave a Reply

Your email address will not be published. Required fields are marked *