[ad_1]
The Bongard issues on the 1 aspect of the Meta-Bongard are all straightforward to unravel (Numbers are OP’s reference system):
1A:
Nothing | One thing
1B:
White form | Black form
1C:
Sq. | Non-Sq. Rectangle
1D:
Large factor | Small factor
1E:
Threesome | Foursome
1F:
Black Circle | White Triangle
The 2 aspect is considerably tougher… (Daring Numbers are Meta-Bongard field, Italic are Bongard)
2A:
Right here, the 2 sides are precisely the identical, so there is no such thing as a potential rule to distinguish them.
2B:
Right here, there may be an obvious rule: Organized | Scrambled.
Sadly, the order of the items can’t be used to differentiate the perimeters of a Bongard drawback, and the items themselves are in any other case an identical.
2C:
This one is identical as 1C – besides that its 2D has been changed with a Sq., breaking the rule and stopping a brand new rule from being potential (because the shapes are in any other case an identical).
2D:
Right here, the 2 sides are virtually, however not fairly, an identical – shades of 2A! There may be an obvious rule right here, although: Greater | Smaller.
Sadly, very like in 2B, this type of comparability is not a sound method to make a rule, and the “small” issues are too shut in dimension to the “huge” issues to only copy-paste the rule for 1D.
2E:
Right here, each side are only a single dot, randomly positioned. There’s virtually a rule right here (Centered | Not Centered), however 2F breaks it.
2F:
Right here, each side are simply noise. I am certain a devoted pc might tease a rule out of this one, but it surely will not be human readable and can really feel very contrived – greatest to not trouble.
So what is the rule for the Meta-Bongard?
I remorse to tell you that I cheated – I decoded @athin’s rot13 remark beneath the query. The rule is Legitimate | Invalid.
That’s, the 1 aspect are all legitimate Bongard issues, and the 2 aspect are all invalid Bongard issues. I had an inkling of this, however 2C threw me for a loop (I could not see that it was completely different from 1C till it was too late).
[ad_2]
Source_link