Skip to main content
9 events
when toggle format what by license comment
Apr 26, 2020 at 6:05 comment added joro @WillSawin May I have your opinion about abc and surjective polynomials: mathoverflow.net/questions/332189/…
Apr 25, 2020 at 16:03 comment added Wolfgang @SylvainJULIEN also note that the threshold of 1.4 used here is completely arbitrary...
Apr 25, 2020 at 15:42 history edited joro CC BY-SA 4.0
Minor fix
Apr 25, 2020 at 14:42 comment added Will Sawin If one assumes $\operatorname{rad}(a), \operatorname{rad}(b), \operatorname{rad}(a+b)$ behave as independent random variables, the key quantity is understanding the number of $n \approx N$ with $\operatorname{rad}(n) \approx n^\alpha$. Surely this must be known/studied. I think the highest-order terms in the asymptotic are $n^\alpha e^{ \sqrt{\log n}}$ but I don't know what the next ones are.
Apr 25, 2020 at 14:07 comment added joro @WillSawin Yes. Or in other words quality >1.
Apr 25, 2020 at 14:00 comment added Will Sawin "good" = $\operatorname{rad}(abc)<c$?
Apr 25, 2020 at 13:08 comment added joro @SylvainJULIEN I strongly doubt this holds in general, we are examining very small set.
Apr 25, 2020 at 12:53 comment added Sylvain JULIEN 58% is not far from $\gamma$.
Apr 25, 2020 at 12:23 history asked joro CC BY-SA 4.0