Network To System
What is the difference between a graph and a system? A graph is a formal mathematical structure that can model a given system as a network. A system, however, is a somewhat more nebulous concept. It is a distinct whole formed from a network of parts, yet separate from its wider environment.
A graph can be modelled as a system when it has a clearly defined boundary around it that allows the graph to be perceived as a unified whole. Is this like clustering, where connected nodes in a graph are grouped into communities? Not quite—a system is more than that. We need to think about boundaries in greater detail.
According to the Free Energy Principle, the boundary of a system is coupled to its environment in a way that selectively regulates information exchange. This selective coupling helps the system resist entropy, increasing order and minimising uncertainty. Rosen goes further, suggesting that in living systems, the presence of this boundary enables internal Closure To Efficient Causation. Efficient causation, an Aristotelian term, refers to the immediate source of change in a system—the "cause" that directly brings about an effect. In other words, only a system with a boundary can possess agency, autonomy, and self-determination.
The practical implications of this are HUGE! As we move into an increasingly networked world, where vast amounts of data are ingested and compressed into foundational AI models, establishing an 'Efficient Causation Boundary' around your organisation is vital. Organisations already have a form of boundary with their firewalls and data leakage policies, but these are woefully inadequate for operating in the 'Age of AI'. The message is stark and clear—establish your information boundary or lose your autonomy and agency!
So how can this be done, and can graphs and AI help?
There is a sense in which connectivity is a precursor to boundary formation; the network must be clustered together to be contained within the boundary. Equally true, once a boundary is established, connectivity increases within the boundary. From this, organisations can draw their first practical lesson:
🔵 Identify your private data and connect it together as tightly as possible.
There is also a sense that if one has a highly connected cluster, there will be a central point around which the cluster is formed—a centroid. The nucleus of a cell, the core values of a nation, centroids like these bind the collective together. From this, organisations can draw their second practical lesson:
🔵 Identify the unique concepts that define your organisation and cluster your data around those concepts.
Graphs and AI can help with these tasks but they can't do it alone. This is a human effort—one that requires the whole organisation to come together, connect its data, discover its Ontological Core, and build a boundary that will allow it to survive in the Age of AI.
⭕ Ontological Core: https://www.knowledge-graph-guys.com/blog/your-ontological-core
