4 ms·
To be fair, results like this are very improbable to come from someone who is not a full-time mathematician. And no that does not mean someone working in machin
by distcs 4y ago
To be fair, results like this are very improbable to come from someone who is not a full-time mathematician. And no that does not mean someone working in machine learning or someone who did their PhD 6 years ago. How many such results do you see from someone who is not a full-time mathematician?
It is possible for someone who is not a full-time mathematician by still working on mathematics in their spare time or in a mathy-field in their day job but very very few real examples of that can be found.
So I don't find the title clickbaity. It is really really hard to come up with results like this without working on mathematics research (not AI research) 10 hours a day!
- jamincan 4y agoThe article also notes that other mathematicians very quickly advanced his work from 10% to 38% and wondered why he hadn't just done it himself, since it was relatively simple. He explained that his skills were rusty after being out of mathematics for six years.
- jebarker 4y agoTotally agree. I did a PhD in Pure Math. 6 years later I could no longer give anything but a superficial explanation of my own thesis. Now ~15 years later my thesis is totally impenetrable to me. Advanced math is a house of cards that quickly falls if you're not actively reinforcing it.