Is concept expressed in "Autodidactic Universe" article plausible? As I understood, the authors (Lee Smolin et al) of the "Autodidactic Universe" article suggest that the fundamental laws of nature as time progresses since the Big Bang event (which happened about 13.8 billion years ago) were/(perhaps still are) undergoing the “learning” self change adoptive process that is somewhat similar to the Charles Darwin's life evolution theory.
Is concept expressed in the"Autodidactic Universe" article plausible?
 A: There's less to pop physics than meets the eye. The full technical content of the argument is:

*

*Physical theories are described by sets of complicated differential equation.

*Training a neural network can be described as a set of complicated differential equation.

*Since you can get any differential equations you want by choosing an arbitrarily complicated physical theory, you can choose one that looks like training a neural network.

*Therefore, the universe is a neural network.

If this seems a bit confusing, note that it's precisely the same as this argument:

*

*In physics, particles have masses, which are numbers.

*In my grocery store, a blueberry pie costs \$15.

*Suppose there exists a new particle whose mass is 15 electron-volts.

*Therefore, since $15 = 15$, the universe is blueberry pie.

Now, is this pie-universe theory plausible? It probably depends on the details of the new particle. But does it actually establish any kind of relation to pie? No, it just means the authors forced a relation in their setup. Most new physical theories you read about in pop physics, like Wolfram's and Weinstein's, are like this.
A: *

*The state of the Universe is governed by a function.


*Neural networks approximate functions -- it is what they are designed to do. Period.


*Therefore, the Universe is an approximation of some ideal and unattainable function ... wait, what? Why? Why bother? Why do we need an approximation when the real thing will do? There is an acronym -- KISS -- keep it simple, stupid.
The only reason they use the word autodidactic is so that they can skip the question of who's doing the training -- God? Ancient astronauts?
It is the cherry picking of concepts from computer science, is what it is. Nowhere do they mention the use of the logistic activation function, or activation functions in general. Weird.
In any case, they're years behind in their research. The latest and greatest toy is the Convolutional Neural Network (CNN). For instance, dog vs cat classification fails 50% of the time using a traditional neural network. The failure rate is 10-20% for the corresponding CNN. They only mention CNNs in passing. In other words, their model is out-of-date out-of-the-gate.
