1
00:00:00,435 --> 00:00:02,213
Choose any positive whole number.

2
00:00:03,104 --> 00:00:05,088
If it is even, divide by two.

3
00:00:05,978 --> 00:00:08,774
If it is odd, multiply by three and add one.

4
00:00:09,663 --> 00:00:10,333
Repeat.

5
00:00:11,222 --> 00:00:13,774
Starting at six, the path reaches one.

6
00:00:14,663 --> 00:00:18,581
The Collatz conjecture says this happens for every positive starting number.

7
00:00:19,470 --> 00:00:20,462
A simple rule.

8
00:00:21,351 --> 00:00:22,614
A universal claim.

9
00:00:24,560 --> 00:00:26,660
A number can rise before it falls.

10
00:00:27,551 --> 00:00:31,610
Checking a huge collection of starts still leaves infinitely many unchecked.

11
00:00:32,500 --> 00:00:36,971
A proof must exclude every alternative: an endless trajectory that escapes to

12
00:00:36,984 --> 00:00:40,579
infinity, and a cycle other than four, two, one.

13
00:00:41,469 --> 00:00:44,613
A convincing pattern is not yet a universal argument.

14
00:00:46,518 --> 00:00:47,936
Why spend effort on this?

15
00:00:48,826 --> 00:00:53,234
Collatz tests how much we understand about deterministic arithmetic dynamics:

16
00:00:53,440 --> 00:00:56,339
exact rules with complicated long term behavior.

17
00:00:57,229 --> 00:01:00,991
Progress connects number theory, probability, and the study of when

18
00:01:01,004 --> 00:01:02,280
algorithms terminate.

19
00:01:03,169 --> 00:01:05,888
Its value is in understanding and new methods.

20
00:01:06,777 --> 00:01:09,625
A practical payoff cannot be promised in advance.

21
00:01:11,560 --> 00:01:14,872
The problem is traditionally associated with Lothar Collatz.

22
00:01:15,762 --> 00:01:19,757
In the nineteen seventies, Riho Terras and, independently, C.

23
00:01:19,770 --> 00:01:20,041
J.

24
00:01:20,054 --> 00:01:24,023
Everett proved that almost every start eventually falls below itself, in

25
00:01:24,036 --> 00:01:25,041
natural density.

26
00:01:25,930 --> 00:01:30,144
Daniel Bernstein and Jeffrey Lagarias developed a precise two adic coding of

27
00:01:30,157 --> 00:01:31,059
the dynamics.

28
00:01:31,948 --> 00:01:35,170
These are different advances toward understanding the same problem.

29
00:01:37,102 --> 00:01:40,980
Terence Tao made a major advance in twenty nineteen, published in twenty

30
00:01:40,993 --> 00:01:41,779
twenty two.

31
00:01:42,670 --> 00:01:45,930
Choose any bound that tends to infinity, however slowly.

32
00:01:46,820 --> 00:01:51,240
Almost every orbit eventually goes below that bound, in logarithmic density.

33
00:01:52,129 --> 00:01:54,577
This is much stronger control of orbit minima.

34
00:01:55,466 --> 00:01:57,941
But it does not say that every orbit reaches one.

35
00:01:58,830 --> 00:02:01,678
The exceptional starts remain the central difficulty.

36
00:02:03,602 --> 00:02:07,648
Our contribution is an evolving research map and a collection of internally

37
00:02:07,661 --> 00:02:08,756
reviewed arguments.

38
00:02:09,647 --> 00:02:13,513
The current snapshot has three hundred fifty six nodes and five hundred

39
00:02:13,526 --> 00:02:15,072
thirty two relationships.

40
00:02:15,961 --> 00:02:18,539
Green means checked within the stated assumptions.

41
00:02:19,428 --> 00:02:21,631
Unfilled nodes mark open obligations.

42
00:02:22,521 --> 00:02:25,820
The count of green nodes measures neither the percentage solved nor the

43
00:02:25,832 --> 00:02:27,250
likelihood of success.

44
00:02:28,139 --> 00:02:31,335
External review and novelty remain separate questions.

45
00:02:33,268 --> 00:02:37,263
In one restricted family, recent work separates a difficult denominator

46
00:02:37,276 --> 00:02:42,121
estimate into two sources: deep divisibility and large multiplicative orders.

47
00:02:43,012 --> 00:02:46,930
A further reduction compresses a first coefficient certificate to degree

48
00:02:46,942 --> 00:02:51,015
below four d, independent of the full matrix length, under stated

49
00:02:51,027 --> 00:02:51,904
assumptions.

50
00:02:52,793 --> 00:02:56,105
The needed bounds for the actual coefficients are still missing.

51
00:02:56,994 --> 00:03:01,027
Even closing this branch would not settle arbitrary Collatz trajectories.

52
00:03:02,935 --> 00:03:07,355
This map and explanation are by Khamit Kadyrbekov and Daniyal Kadirbekov,

53
00:03:07,600 --> 00:03:11,762
with artificial intelligence assistance for exploration, drafting, and

54
00:03:11,775 --> 00:03:12,845
internal checking.

55
00:03:13,735 --> 00:03:17,988
Our workflow separates arithmetic derivation, structural reformulation,

56
00:03:18,143 --> 00:03:20,836
critical review, and connection to the full goal.

57
00:03:21,725 --> 00:03:24,895
Earlier mathematicians are credited for their published work.

58
00:03:25,784 --> 00:03:28,027
They are not collaborators on this project.

59
00:03:28,916 --> 00:03:31,867
Responsibility for the claims remains with the authors.

60
00:03:33,768 --> 00:03:37,402
One route to a full proof is to show that every odd start greater than one

61
00:03:37,415 --> 00:03:39,232
eventually falls below itself.

62
00:03:40,123 --> 00:03:42,649
Strong induction would then finish the argument.

63
00:03:43,538 --> 00:03:46,038
We have not established that universal descent.

64
00:03:46,927 --> 00:03:50,084
There is no defensible completion date or percentage remaining.

65
00:03:50,973 --> 00:03:55,432
Explore the map, inspect the assumptions, challenge a step, or help prove a

66
00:03:55,445 --> 00:03:57,133
clearly stated missing lemma.

67
00:03:58,022 --> 00:04:00,303
That is how this project can move forward.