News for hackers. Stuff that matters.
Re:Researchers chip away at Smale's 7th unsolved problem in ... by paulpauper on Saturday July 16, 2016 @02:57PM (#12107461)
but Wikipedia shows a huge table of values.it's just that we don't have a name or certain constructions for values above 12
Re:Researchers chip away at Smale's 7th unsolved problem in ... by Jabbles on Saturday July 16, 2016 @03:30PM (#12107567)
Those are the best known configurations, we don't have proofs that they are the lowest possible.
Re:Researchers chip away at Smale's 7th unsolved problem in ... by paulpauper on Saturday July 16, 2016 @03:34PM (#12107579)
Isn't this an optimization problem that can be solved with computer to solve a high degree equation?
Re:Researchers chip away at Smale's 7th unsolved problem in ... by jessriedel on Saturday July 16, 2016 @05:23PM (#12107927)
Not sure. For n points on a sphere, you're trying to prove a global minimum a very bumpy function over 2n-dimensional space. It may become computationally intractable.
Re:Researchers chip away at Smale's 7th unsolved problem in ... by Someone on Saturday July 16, 2016 @05:56PM (#12108032)
And even if you can find a numerical solution to arbitrary precision, at best you only get arbitrarily close to the mathematically optimal solution (at best because a purely numerical approach cannot rule out that it misses a very small, but high peak in the function to be optimized)

As an extreme example, for n=2 and using binary floating point, you probably will find the mathematically correct solution, but there's no way to tell numerically whether your answer is off by a fraction of your floating point's epsilon value.

That's probably not important to physicists who want to know an answer, but it is for mathematicians.

What I find very surprising is (from https://en.m.wikipedia.org/wiki/Thomson_problem):

"Numerical solutions for N=8 and 20 are not the regular convex polyhedral configurations of the remaining two Platonic solids, whose faces are square and pentagonal, respectively."

I would like to see the visualizations of the better solutions for n=20 (wikipedia links to the one for n=8) and/or hear a heuristic argument as to how that can happen.

Re:Researchers chip away at Smale's 7th unsolved problem in ... by grkvlt on Saturday July 16, 2016 @06:13PM (#12108081)
For N=8 the solution listed on the Wikipedia page is a square antiprism (https://en.m.wikipedia.org/wiki/Square_antiprism) which seems intuitively reasonable. But, I don't see anything listed for N=20. Look at the table in S.5 of the article (https://en.m.wikipedia.org/wiki/Thomson_problem#Configuratio...) for more links to visualisations in the rightmost column, if one exists.
Re:Researchers chip away at Smale's 7th unsolved problem in ... by tamana on Saturday July 16, 2016 @07:34PM (#12108304)
The function here has a simple continuity, so there is no risk of narrow peaks beyond some computable threshold. The challenge is the high dimensionality.
Re:Researchers chip away at Smale's 7th unsolved problem in ... by contravariant on Saturday July 16, 2016 @06:55PM (#12108190)
The unsolved problem referred to in the title is finding a configuration close to the global minimum. Most optimization algorithms need such a configuration as a starting point.