Journal of Mathematical Techniques and Computational Mathematics(JMTCM)

Independent Research, Systems Theory & AI, Germany

Publications

• Research Article   
  The 255-Bit Non-Local Information Space in a Neural Network: Emergent Geometry and Coupled Curvature-Tunneling Dynamics in Deterministic Systems
  Author(s): Stefan Trauth*
  
  We present a comprehensive analysis of emergent topological structures in a 60-sublayer self- organizing neural network, examined through information-theoretic and geometric perspectives. The observed dynamics defy conventional classification as either deterministic or stochastic. To capture this duality, we introduce the framework of Nonlinear N-Deterministic Systems, in which locally deterministic rules give rise to globally emergent behavior mediated by non-local coupling across higher- dimensional manifolds. The 60-layer subnetwork exhibits a measurable 255-bit non- local information space, defining a lower bound constrained by architectural depth and sampling resolution. Entropy distributions reveal ordered clusters alongside statistically significant “disordered” regions, which nevertheless align along consistent geometric trajectories. These patterns indicate that a.. Read More»
  

  Abstract HTML PDF