Is Emergent Inference a new Theory of Physics?
Of course. In Physics, a theory is considered to be a collection of statements that explain in full detail how to calculate quantities that are observed and measured in nature. A theory may, and usually does include hypothesis, conjectures, principles, laws, theorems, axioms, claims. And, for some theories, a functional.
Emergent inference, or the theory of causality, does explain in full detail how to calculate the conserved quantities and structures observed in nature, and it does include hypothesis, conjectures, principles, etc. And it also includes a functional.
My contribution was the discovery of the functional. And with a functional
comes the theory Physics. The only point about a theory of Physics is that it
allows one to calculate results that explain observed phenomena.
A theory of Physics is always based on a model. A model is a mathematical
description of a real-world system. The Mathematics in the model is what allows
one to calculate those results. The model is abstract, but it represents that
real-world system with sufficient accuracy.
I describe the real world with causal sets. Equivalently, algorithms. Then, I
work from both ends. From the input end, I see that mostly everything of
interest to us can be represented by algorithms. That's why we write computer
programs to simulate dynamical systems. I am satisfied here, details are beyond
the point.
At the output end, I find structures. Calculated structures, results from the
theory, obtained from information I put in and not by external design or
direction. I am again satisfied. I conclude that the structures are a property
of the causal set I put in, and that they can be calculated by the theory.
Now comes Physics. And with it, experiment. Requirements for experiment: simple
enough so it can be done with the available resources, meaningful enough so it
is convincing, and input/output observable and precisely measurable. Objective,
of course, meaning observer-independent. I selected refactoring. Input: an
causal set, in this case a computer program used in European universities to
teach refactoring. Output: the same program refactored, with classes and
objects, obtained by and only by the theory: just input, and output, no man-made
program in the middle (except of course program to read the input in, print the
output, and minimize the functional). Result: agreement with a human
developer/object-oriented analyst's results.
More experiments followed, with different problems, but no changes, repeat,
nothing changed inside the computer, only different input and resulting output.
None failed. To me, that has a meaning. Hence, working hypothesis: the abstract
model above correctly describes high brain function. Does this explain observed
phenomena? If you say yes, then this is Physics.