You're reading: Posts Tagged: proof news

P might not be NP, reckons Norbert Blum

Norbert Blum of Universität Bonn has uploaded to the arXiv a preprint of a paper claiming to resolve the problem of whether $\mathrm{P} = \mathrm{NP}$, in the negative.

“Proofs” one way or the other turn up on the arXiv pretty much every day, but this one might actually be correct. At least, it’s not immediately obvious it isn’t.

Here’s the abstract:

Berg and Ulfberg and Amano and Maruoka have used CNF-DNF-approximators to prove exponential lower bounds for the monotone network complexity of the clique function and of Andreev’s function. We show that these approximators can be used to prove the same lower bound for their non-monotone network complexity. This implies $\mathrm{P} \neq \mathrm{NP}$.

John Baez has very quickly put together a post explaining the very basics of Blum’s argument.  Even more briefly, Blum claims to have shown that the best-case complexity of a function solving the clique decision problem is exponential, not polynomial.

Colin Wright reckons that the proof passes all of Scott Aaronson’s immediate ‘sniff tests’ for a claimed proof of a big problem, and his supplementary list for proofs to do with P versus NP. Those help you spot charlatans and Walter Mitty types, rather than looking at the actual mathematical content.

Obviously, none of us are qualified to even offer a hot take on this, so we’ll all have to wait until more experienced sorts have had a good look.

So, watch this space.

(Personally, my money is on this not quite working, purely based on my natural pessimism)

Not mentioned on The Aperiodical, March 2016

There’s been a lot of maths news this month, but we’ve all been too busy to keep up with it. So, in case you missed anything, here’s a summary of the biggest stories this month. We’ve got two new facts about primes, the best way of packing spheres in lots of dimensions, and the ongoing debate about the place of maths in society, as well as the place of society in maths.

A surprisingly simple pattern in the primes

Kannan Soundararajan and Robert Lemke Oliver have noticed that the last digits of adjacent prime numbers aren’t uniformly distributed – if one prime ends in a 1, for example, the next prime number is less likely to end in a 1 than another odd digit. Top maths journos Evelyn Lamb and Erica Klarreich have both written very accessible pieces about this, in the Nature blog and Quanta magazine, respectively.

Oliver and Soundararajan’s paper on the discovery is titled “Unexpected biases in the distribution of consecutive primes”.

László Babai reckons he can decide if two graphs are isomorphic in quasipolynomial time

László Babai in Chicago. Photo by Gabe Gaster, used with permission.

We’ve been slow to cover this, but if this week has taught us anything, it’s that taking your time over Important Maths News is always a good idea.

A couple of weeks ago, rumours started circulating around the cooler parts of the internet that László “Laci” Babai had come up with an algorithm to decide if two graphs are isomorphic in quasipolynomial time. A trio of mathematicians including Tim Gowers were on BBC Radio 4’s In Our Time discussing P vs NP while these rumours were circulating and made a big impression on Melvyn Bragg as they talked so excitedly about the prospect of something big being announced.

If Babai had done what the rumours were saying, this would be a huge advance – graph isomorphism is known to be an NP problem, so each step closer to a polynomial-time algorithm raises the P=NP excite-o-meter another notch.

Erdős’s discrepancy problem now less of a problem

Boris Konev and Alexei Lisitsa of the University of Liverpool have been looking at sequences of $+1$s and $-1$s, and have shown using an SAT-solver-based proof that every sequence of $1161$ or more elements has a subsequence which sums to at least $2$. This extends the existing long-known result that every such sequence of $12$ or more elements has a subsequence which sums to at least $1$, and constitutes a proof of Erdős’s discrepancy problem for $C \leq 2$.