3 ms·
If you can generalize on similar problems and be correct more often than random chance, then you have a useful model regardless of the data backing that model.
by lightcatcher 11y ago
If you can generalize on similar problems and be correct more often than random chance, then you have a useful model regardless of the data backing that model.