# Thread: Ernie and random numbers

1. So in other words, relating all this to the op which is the whole point, let's say you got Ernie to generate 1e44 random numbers. I don't know how long that would take, probably a long time haha. But let's say you could do it.

The quantum theory says, that after 1e44 samples your observed statistical distribution will be exactly perfect.

However my claim is, it will be imperfect. You will see some deviations from Pure Randomness, and you will have to account for those. Most likely the accounting will occur on the basis of information and geometry.

So, the testing in the OP which is being done to four or five orders of magnitude, is completely inadequate. It needs to be done to 44 orders of magnitude, not 4. But I mean, this is a wonderful thing, having machines which can finally do this.

If my thinking is correct, geometry is information to. If you're dividing something into two parts, you're doing work. The work that you're doing ends up as a partition, which is to say geometry. A partition is the same thing when you divide a set, or when you choose a member from a set.

This concept underlies the field of "information geometry". (Google Shun-Ichi Amari).  Reply With Quote

2. Originally Posted by nonsqtr So in other words, relating all this to the op which is the whole point, let's say you got Ernie to generate 1e44 random numbers. I don't know how long that would take, probably a long time haha. But let's say you could do it.

The quantum theory says, that after 1e44 samples your observed statistical distribution will be exactly perfect.

However my claim is, it will be imperfect. You will see some deviations from Pure Randomness, and you will have to account for those. Most likely the accounting will occur on the basis of information and geometry.

So, the testing in the OP which is being done to four or five orders of magnitude, is completely inadequate. It needs to be done to 44 orders of magnitude, not 4. But I mean, this is a wonderful thing, having machines which can finally do this.

If my thinking is correct, geometry is information to. If you're dividing something into two parts, you're doing work. The work that you're doing ends up as a partition, which is to say geometry. A partition is the same thing when you divide a set, or when you choose a member from a set.

This concept underlies the field of "information geometry". (Google Shun-Ichi Amari).
Why? very very few things in physics require 44 orders of magnitude accuracy.

Fine structure constant - 8 decimal places
gauge coupling constant - 8 decimal places

Quark masses - 4 decimal places (in MeV or GeV)
CMK mixing angles (x4) - 2 decimal places
W Z and Higs boson massess - 4 dec places max.
Neutrino masses - 2 decimal places in eV
Proton to Electron mass ratio - 10 dec places

Fact is even on amicroscopic scale, not much needs more than 10 dcimal places, in Si units. why do you think you need 44 orders of magnitude to determine randomness. ERNIE's sample of 8.5 million numbers is easily statistically significant.

This

ERNIE's activities are also checked before and after each draw by the Government Actuary's Department (GAD), which performs four tests to check the machine is truly random:

• Frequency – whether every possible character in each position of the Bond number appears as often as it should
• Serial – this looks at how many times one digit follows another, e.g. how often does a 3 directly follow a 7
• Poker – this looks at the number of times consecutively generated groups of characters contain: four identical characters, three of a kind, two pairs, one pair, all different
• Correlation – this looks for correlation between characters in two different Bond positions over a series of Bond numbers

ERNIE has never failed.

This method passes every known mathematical method for evaluating randomness  Reply With Quote

3. Regarding the question of heterogeneous entanglement:

https://m.phys.org/news/2018-06-scie...antum-dot.html  Reply With Quote