Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Hi, I read the post. The numerical approach gives the data. But the proof of the result uses analysis. I was talking about the proof. Thanks.


The roots of a polynomial vary smoothly as the coefficients vary. This is intuitively obvious. This is an important part of Abel's proof of the unsolvability of the quinitic. The probability of a random number being ambiguously real or complex has measure 0 (measure < epsilon, for all epsilon > 0), so you can ignore all computationally ambiguous cases. You don't need an exact calculation of any specific polynomial's root, to get the right answer of the measurement over all polynomials.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: