Notes from the search for general intelligence
From the archive
The brain is the only existence proof we have that general intelligence is possible. I am trying to understand how it works and build something inspired by what I find. These are the notes from that search.
The Thinking Archive is a record of one researcher's search for the principles underlying general intelligence.
I am Amritansh, an independent AI researcher based in Bengaluru, India, working at the intersection of computational neuroscience, mathematics, physics, and machine learning.
What I have is a deep conviction that the brain — the only existence proof of general intelligence we possess — contains mathematical principles that current AI has not yet formalised. My research attempts to extract those principles and build with them.
This archive exists because research produces more than papers. It produces failed experiments, half-formed ideas, unexpected connections, daydreams that turn out to matter. Everything here is published openly, for anyone who finds it useful.
The archive does not forget. That is the point.
"The brain is not a computer that happens to be made of meat. It is a dynamical system that has learned, over five hundred million years, how to stay on the edge of a phase transition — just organised enough to act, just disordered enough to adapt."
Current AI optimises loss functions.
Biological intelligence navigates bifurcations.
These are not the same problem.
They do not have the same solution.
My research is an attempt to close that gap —
formalising the brain's plasticity-stability mechanism
as a mathematical framework for machines that learn
without forgetting.