Unbiased

Post puzzles for others to solve here.

Unbiased

Postby Yogi » Mon Oct 11, 2021 10:21 am

12.74.6.94..8......89..3.1....4.2..1..........4.6..9..3...2...62....6.....8...4..
Code: Select all
+---+---+---+
|12.|74.|6.9|
|4..|8..|...|
|.89|..3|.1.|
+---+---+---+
|...|4.2|..1|
|...|...|...|
|.4.|6..|9..|
+---+---+---+
|3..|.2.|..6|
|2..|..6|...|
|..8|...|4..|
+---+---+---+

What would the term Unbiased mean in Sudoku?
User avatar
Yogi
2017 Supporter
 
Posts: 339
Joined: 05 December 2015
Location: New Zealand

Re: Unbiased

Postby denis_berthier » Mon Oct 11, 2021 11:14 am

.
Code: Select all
Resolution state after Singles and whips[1]:
   +----------------------+----------------------+----------------------+
   ! 1      2      3      ! 7      4      5      ! 6      8      9      !
   ! 4      567    567    ! 8      19     19     ! 2357   2357   2357   !
   ! 57     8      9      ! 2      6      3      ! 57     1      4      !
   +----------------------+----------------------+----------------------+
   ! 8      3579   57     ! 4      3579   2      ! 357    6      1      !
   ! 5679   135679 12567  ! 1359   13579  1789   ! 2357   4      23578  !
   ! 57     4      1257   ! 6      1357   178    ! 9      2357   23578  !
   +----------------------+----------------------+----------------------+
   ! 3      1579   157    ! 159    2      4      ! 8      579    6      !
   ! 2      579    4      ! 359    8      6      ! 1      3579   357    !
   ! 569    1569   8      ! 1359   13579  179    ! 4      2359   235    !
   +----------------------+----------------------+----------------------+
150 candidates.


Solution with very elementary steps - nothing more complicated than bivalue-chains[2]:
Code: Select all
naked-pairs-in-a-block: b4{r4c3 r6c1}{n5 n7} ==> r6c3≠7, r6c3≠5, r5c3≠7, r5c3≠5, r5c2≠7, r5c2≠5, r5c1≠7, r5c1≠5, r4c2≠7, r4c2≠5
naked-pairs-in-a-column: c1{r3 r6}{n5 n7} ==> r9c1≠5
finned-x-wing-in-columns: n9{c1 c4}{r5 r9} ==> r9c6≠9, r9c5≠9
whip[1]: b8n9{r9c4 .} ==> r5c4≠9
biv-chain[2]: r3n5{c7 c1} - b4n5{r6c1 r4c3} ==> r4c7≠5
biv-chain[2]: r3n7{c7 c1} - b4n7{r6c1 r4c3} ==> r4c7≠7
singles ==> r4c7=3, r4c2=9, r5c1=6, r9c1=9, r2c3=6, r9c2=6, r5c2=3, r7c2=1, r6c5=3
finned-x-wing-in-rows: n5{r3 r6}{c1 c7} ==> r5c7≠5
whip[1]: c7n5{r3 .} ==> r2c8≠5, r2c9≠5
biv-chain[2]: r7n7{c8 c3} - c2n7{r8 r2} ==> r2c8≠7
biv-chain[2]: r7n7{c8 c3} - b4n7{r4c3 r6c1} ==> r6c8≠7
whip[1]: c8n7{r8 .} ==> r8c9≠7
hidden-pairs-in-a-block: b9{n7 n9}{r7c8 r8c8} ==> r8c8≠5, r8c8≠3, r7c8≠5
finned-x-wing-in-rows: n5{r4 r7}{c3 c5} ==> r9c5≠5
whip[1]: c5n5{r5 .} ==> r5c4≠5
stte



Yogi wrote:What would the term Unbiased mean in Sudoku?

In proba and stats, given a probability distribution, it applies to an estimator of some parameter (e.g. the mean...): an estimator (which is a random variable) is said unbiased if its mean value is equal to the true value of the distribution.

In Sudoku, it is generally supposed to apply to the number of clues and there are two contexts in which it occurs:
- complete grids: as we know the full collection of complete grids, it's quite easy to have an unbiased estimator.
- minimal puzzles: all the known generators are strongly biased wrt to the number of clues (but more or less). See [PBCS, chapter 6] for a comparison of different types of generators). The most biased generators are bottom-up. The least biased known generator is the controlled_bias generator (https://github.com/denis-berthier/Controlled-bias_Sudoku_generator_and_collection)


Why are you asking this question about this puzzle?
denis_berthier
2010 Supporter
 
Posts: 3982
Joined: 19 June 2007
Location: Paris

Re: Unbiased

Postby Yogi » Mon Oct 11, 2021 8:46 pm

Thank you. Your answer indicates that the term unbiased is not important to me.
The puzzle is an example which came from a huge document, downloaded from somewhere, containing a so-far unknown number of codes.
The document can be read by MS Word but I don't think it is in an up-to-date format. The puzzles are of varying difficulty and I thought that one was reasonably interesting. The document was named Unbiased.
User avatar
Yogi
2017 Supporter
 
Posts: 339
Joined: 05 December 2015
Location: New Zealand

Re: Unbiased

Postby eleven » Mon Oct 11, 2021 9:19 pm

Done with fun with 57.
Code: Select all
+----------------------+----------------------+----------------------+
| 1      2      3      | 7      4      5      | 6      8      9      |
| 4      567    567    | 8      19     19     | 2357   2357   2357   |
|#57     8      9      | 2      6      3      |#57     1      4      |
+----------------------+----------------------+----------------------+
| 8      39    #57     | 4      3579   2      | 3-57   6      1      |
| 69     1369   126    | 1359   13579  1789   | 2357   4      23578  |
|#57     4      12     | 6      1357   178    | 9      2357   23578  |
+----------------------+----------------------+----------------------+
| 3      1579   157    | 159    2      4      | 8      579    6      |
| 2      579    4      | 359    8      6      | 1      3579   357    |
| 69     1569   8      | 1359   13579  179    | 4      2359   235    |
+----------------------+----------------------+----------------------+

Remote pair r3c71,b4p73 => -57r4c7

Singles and 9 locked candidate ...
Code: Select all
+-------------------+-------------------+-------------------+
| 1     2     3     | 7     4     5     | 6     8     9     |
| 4     57    6     | 8     19    19    | 257   2357  2357  |
|#57    8     9     | 2     6     3     |#57    1     4     |
+-------------------+-------------------+-------------------+
| 8     9     57    | 4     57    2     | 3     6     1     |
| 6     3     12    | 15    1579  1789  |#57+2  4     578-2 |
|#57    4     12    | 6     3     178   | 9    #57+2 @578-2 |
+-------------------+-------------------+-------------------+
| 3     1     57    | 59    2     4     | 8     579   6     |
| 2     57    4     | 359   8     6     | 1     3579  357   |
| 9     6     8     | 135   157   17    | 4     235   235   |
+-------------------+-------------------+-------------------+

Oddagon 57 in r63c1, r35c7, r6c8 has 2 guardians 2b6p48 => -2r56c9
Then oddagon 57 in r63c1, r35c7, r6c9 has 2 guardians 2r5c7,8r6c9:
8r6c9 == 2r5c7 - (2=578)r6c189 => 8r6c9

Some singles, 7 locked candidate ..
Code: Select all
+-------------------+-------------------+-------------------+
| 1     2     3     | 7     4     5     | 6     8     9     |
| 4     57    6     | 8     1     9     | 257   2357  2357  |
| 57    8     9     | 2     6     3     | 57    1     4     |
+-------------------+-------------------+-------------------+
| 8     9    #57    | 4    #57    2     | 3     6     1     |
| 6     3     12    | 15    9     8     | 257   4     57    |
| 57    4     12    | 6     3     17    | 9     25    8     |
+-------------------+-------------------+-------------------+
| 3     1    #57    | 9-5   2     4     | 8     579   6     |
| 2     57    4     | 359   8     6     | 1     3579  357   |
| 9     6     8     | 135  #57    17    | 4     235   235   |
+-------------------+-------------------+-------------------+

Remote pair 57 in r74c3,r49c5 => -5r7c4, stte
eleven
 
Posts: 3105
Joined: 10 February 2008

Re: Unbiased

Postby denis_berthier » Tue Oct 12, 2021 1:03 am

Yogi wrote:Thank you. Your answer indicates that the term unbiased is not important to me.
The puzzle is an example which came from a huge document, downloaded from somewhere, containing a so-far unknown number of codes.
The document can be read by MS Word but I don't think it is in an up-to-date format. The puzzles are of varying difficulty and I thought that one was reasonably interesting. The document was named Unbiased.


If it was named unbiased, it's probably cheating on names. There's no unbiased generator and no unbiased collection of puzzles.
denis_berthier
2010 Supporter
 
Posts: 3982
Joined: 19 June 2007
Location: Paris

Re: Unbiased

Postby Yogi » Tue Oct 12, 2021 4:47 am

That's all good. I was taking as probably meaning that it was a totally random collection which has no limiting parameters, such as requiring Symmetry, a certain level of difficulty, number of givens, etc. 'Without preference,' if you like.
User avatar
Yogi
2017 Supporter
 
Posts: 339
Joined: 05 December 2015
Location: New Zealand

Re: Unbiased

Postby denis_berthier » Tue Oct 12, 2021 6:16 am

Yogi wrote:That's all good. I was taking as probably meaning that it was a totally random collection which has no limiting parameters, such as requiring Symmetry, a certain level of difficulty, number of givens, etc. 'Without preference,' if you like.

Unbiased does mean "without preferences" in informal language. Unfortunately, this pseudo "unbiased" collection must have preferences for fewer clues.
denis_berthier
2010 Supporter
 
Posts: 3982
Joined: 19 June 2007
Location: Paris


Return to Puzzles