4 ms·
I would expect to be able to get 100% (on the training data) on this task. Shouldn't any sequences of election victories be easily represented with multiple si
by Imnimo 6y ago
I would expect to be able to get 100% (on the training data) on this task. Shouldn't any sequences of election victories be easily represented with multiple sin functions?
I feel like the message I'm supposed to take from this demo is "Look how good TuringBot is! They can automatically find a function to match this data!" But the actual message I'm getting is "Symbolic regression is too hard for TuringBot!"
- frabert 6y agoIndeed. Just perform an FFT on the data and encode the resulting coefficients in a closed formula.
- bnveg 6y agoThis is a Pareto optimization with a limit on the size of the formulas. Sure, with a formula of arbitrary size you can fit anything with 100% accuracy, but the task is much harder if what you are looking for are short models.