Self Driven Systematic Growth

Computing from the information in front of it

Self Driven Systematic Growth began with a theory: everything is made of something, and what makes it work has an underlying formula or pattern.

We built Self Driven Systematic Growth to investigate whether a model could derive those relationships from the information in front of it and use them to compute, without relying on prior training.

The research brings together three connected questions: how a model discovers a pattern, how it recognises where that pattern applies, and how each discovery can contribute to a growing understanding.

From pattern to computation

Early mathematical experiments demonstrated this process through sequence computation, including successful results on the Fibonacci sequence.

The Fibonacci sequence follows a relationship in which each number is the sum of the two preceding it:

0, 1, 1, 2, 3, 5, 8, 13, 21, 34…

In the recorded tests, Self Driven Systematic Growth derived the relationship from the sequence values presented to it and used that relationship to compute further values correctly. Its answers were checked against values held aside during discovery.

A relationship derived from the information had become a way to compute beyond it.

The mathematical work also recorded successful cases involving Tribonacci and Catalan sequences. These results established concrete examples of pattern derivation and application, providing a foundation for the next stage of the research.

Connecting discoveries

A discovery becomes more useful when a model can recognise where to apply it again.

In related-sequence tests, an established relationship could be recognised from fewer values. Earlier discoveries had become useful in approaching a later encounter.

This is central to the direction of Self Driven Systematic Growth: developing connections between what it has worked out, the situations in which those discoveries apply and the new relationships they help reveal.

We are investigating how these connections can form stronger bonds over time, allowing useful discoveries to contribute to a broader understanding.

Self-driven development

As the research developed, it became apparent that continually understanding and applying patterns required Self Driven Systematic Growth itself to keep evolving.

Each discovery could create further work. Applying a pattern could reveal another relationship, deepen an existing understanding or connect discoveries previously considered separately.

The drive to continue had to come from within the model.

Self Driven Systematic Growth is self-driven. Its activity arises from what it encounters and the work taking place within it. This makes self-driven operation part of how the model develops, with its own discoveries contributing to the work that follows.

The research ahead

The research surrounding Self Driven Systematic Growth continues to explore how derivation, application and connected understanding can develop together.

The mathematical results provide a foundation for investigating more varied information, richer relationships and increasingly demanding situations. The work now studies how useful discoveries remain available, how connections strengthen through further encounters, and how the model’s developing understanding informs what it investigates next.