From the eyes and mind from outside of LODA and OEIS and mathematical sequences...

So what is this all about?!

Sorry, but the present front page description leaves out a few essential details for those that do not already know all about this!


From the front page and the threads, my present guess is:

• We have real world mathematical sequences of integer numbers, and the mathematical formulas for them, listed in an online catalog called OEIS;
  
  
• LODA is a 'simple' computer language that is specially designed/contrived such that it is guaranteed to always complete (no infinite loops for example);
  
  
• LODA programs can be 'automatically' (randomly?) generated to produce a mathematical sequence;
  
  
• This project is to find ("mine") LODA programs that 'happen to' generate integer sequences that match a mathematical sequence listed in OEIS;
  
  
• This project is nothing to do with cryptocurrency mining.
  

Then what?

Why??


Keep searchin'!
Martin

Proposals for improved texts for the website are much welcome.

I'm not a native english speaker, so explaining what LODA is about, in a easy readable way, is hard for me.

Below is my attempt at describing what LODA is.

---

LODA is a math AI, that takes integer numbers as input, and outputs a corresponding formula.

LODA needs your help training the AI. Please join the BOINC community for this.

Is LODA correct?
No, is the short answer.
The long answer, it depends on the training data. The OEIS database is used as training data.

There is math everywhere, such as crystal structures, how many spheres can be fitted in a cube, how quickly can rabbits multiply.
If LODA find a new formula that does things in a smarter way, that what was previously known, it may lead to new discoveries in the field of math/physics. It may reduce power consumption in computers.

Proposals for improved texts for the website are much welcome.

I'm not a native english speaker, so explaining what LODA is about, in a easy readable way, is hard for me.

Below is my attempt at describing what LODA is.

---

LODA is a math AI, that takes integer numbers as input, and outputs a corresponding formula.

LODA needs your help training the AI. Please join the BOINC community for this.

Is LODA correct?
No, is the short answer.
The long answer, it depends on the training data. The OEIS database is used as training data.

There is math everywhere, such as crystal structures, how many spheres can be fitted in a cube, how quickly can rabbits multiply.
If LODA find a new formula that does things in a smarter way, that what was previously known, it may lead to new discoveries in the field of math/physics. It may reduce power consumption in computers.

Really? Is an AI system used to generate and to test the programs/LODA-descriptions?

Or is this a brute force "try all permutations" or "follow a simple generate algorithm" that are typical of a 'mining' brute force search?


Taking your description and this overview:

https://loda-lang.org/
wrote:

LODA is an assembly language, a computational model, and a distributed tool for mining programs. You can use it to generate and search programs that compute integer sequences from the On-Line Encyclopedia of Integer Sequences® (OEIS®). The goal of the project is to find new formulas and more efficient algorithms for a wide range of non-trivial integer sequences.

The LODA Language

LODA means Lexicographical Order Descent Assembly. It is an assembly-based language for solving number-theoretic problems. Its simple syntax combined with its powerful operations makes it ideal for automated search of new programs and algorithms using a process called program mining...

... one possible description might be:


This Boinc project looks for LODA programs ('formulae') that more efficiently generate integer sequences such as the digits of pi, or the list of prime numbers... And many others already known and those yet to be found.

Integer sequences are of great interest in mathematics and they are also a fundamental part of everyday life. We have the Fibonacci sequence, for just one popular example, that can be seen in the optimal arrangement of leaves on a plant stem, or in the arrangement of the seeds in a sunflower, or for how your bones grow. Further beyond life, this sequence is marvelously strung throughout science and human culture and art and beyond. And there are an infinity of other sequences...

The LODA programs that are generated are tested against the On-Line Encyclopedia of Integer Sequences® (OEIS®).

LODA offers methods that may be able to more efficiently generate integer sequences and to allow us to better understand those sequences.

The LODA programs that we find may also be used to speed up generating those sequences so that we can make computers run faster for lower power.

You can help with this project.

Welcome to a deep search!




Keep searchin'
Martin

Martin, your explanation is spot on. I may copy parts of it.

ML algorithms. Currently it's brute force, trying mutating existing programs and see if the candidate program is an entirely new program, or an improvement to an existing program.

In LODA-RUST uses: bloomfilter, pagerank, ngrams, Locality-sensitive_hashing.

This is what the "trigram.csv" with instructions look like

count;word0;word1;word2
20976;lpe;mov;STOP
17672;add;lpe;mov
15456;mov;lpb;sub
14987;START;mov;mov
11889;mov;mov;lpb
11836;lpb;sub;add
11041;lpb;sub;mov
10899;START;mov;add
9100;add;add;add
9094;add;lpb;sub
9062;sub;lpe;mov
8906;START;add;mov
8405;START;mov;lpb
8346;mov;add;lpb
SNIP 4100 rows SNIP

I want to experiment with transformers. I'm not sure how to represent the instructions/registers/values.

Martin, your explanation is spot on. I may copy parts of it.

You're very welcome to copy as is or just parts, no credit needed/wished.

ML algorithms. Currently it's brute force, trying mutating existing programs and see if the candidate program is an entirely new program, or an improvement to an existing program.

In LODA-RUST uses: bloomfilter, pagerank, ngrams, Locality-sensitive_hashing.

This is what the "trigram.csv" with instructions look like...

I want to experiment with transformers. I'm not sure how to represent the instructions/registers/values.

Mmmmm... Machine Learning is not necessarily "Artificial Intelligence" (despite what the popular press get confused with!).

Are you not describing the use of Genetic Algorithms?

Or are you trying multiple heuristics?...

If using a brute force try-all-permutations: What will be the estimated rate of progress? Or is such a simple method good enough to quickly give good results?


Good project!

Thanks,
Martin

If you have a look in your WU logs, you may find how you're contributing: (I think)


your tasks logs => https://boinc.loda-lang.org/loda/logs.php

Search for "faster"

Line  4104: 2022-10-01 01:28:50|ALERT|Faster (IE) program for A095343: Length of n-th string generated by a Kolakoski(7,1) rule starting with a(1)=1. Terms: 1,1,7,7,31,49,145,289,727,1591. Submitted by [AF>EDLS]zOU
	Line  4144: 2022-10-01 02:21:45|ALERT|Faster program for A100821: a(n) = 1 if prime(n) + 2 = prime(n+1), otherwise 0. Terms: 0,1,1,0,1,0,1,0,0,1. Submitted by [AF>EDLS]zOU
	Line  9867: 2022-09-27 10:49:14|ALERT|Faster program for A052102: The second of the three sequences associated with the polynomial x^3 - 2. Terms: 0,1,2,3,6,15,36,81,180,405. Submitted by [AF>EDLS]zOU
	Line 11584: 2022-10-01 17:24:31|ALERT|Faster program for A019733: Decimal expansion of sqrt(2*Pi)/13. Terms: 1,9,2,8,1,7,5,5,9,5. Submitted by [AF>EDLS]zOU
	Line 12004: 2022-10-01 13:53:46|ALERT|Faster program for A053866: Parity of A000203(n), the sum of the divisors of n; a(n) = 1 when n is a square or twice a square, 0 otherwise. Terms: 1,1,0,1,0,0,0,1,1,0. Submitted by [AF>EDLS]zOU
	Line 12339: 2022-10-01 15:16:37|ALERT|Faster program for A049470: Decimal expansion of cos(1). Terms: 5,4,0,3,0,2,3,0,5,8. Submitted by [AF>EDLS]zOU
	Line 12832: 2022-09-28 13:00:57|ALERT|Faster program for A032814: Numbers whose set of base-14 digits is {2,3}. Terms: 2,3,30,31,44,45,422,423,436,437. Submitted by [AF>EDLS]zOU

or maybe I'm just wrong ?
But if I'm right, it would be great to have these discoveries mentioned/counted somewhere or counting for a badge ?