justinlietz93 - Overview

Void Dynamics Model (VDM)

A two part framework built to be attacked:

A background‑independent, metriplectic field theory unifying matter and forces from the Quantum Geometric Tensor, with an emergent causal cone and an epistemological J→M projection.

A real time, zero training, emergent run-time with scale‑free/heavy‑tail sparse neural graphs, locally interacting neuron‑particles (void walkers), Hebbian working memory/online plasticity, sparse activations, and attention‑like state routing driven by physical dynamics.

Lineage: AMN-->FUM-->VDM

DOI ORCID AMD/ROCm only Dual Academic/Commercial License


🔭 What I’m building

  • VDM — Klein-Gordon / reaction–diffusion field intelligence + walker ecology + scoreboard/GDSP gating
  • Brain-Like Cognition — emergent cognition self-organizes from strict fundamental physics constraints and avoidance of traditional ML tactics
  • Memory Steering — dynamic knowledge graph with event-driven updates
  • Agency — inherent, minimal energy optimization rules
  • Real-time control — swap massive pretraining for fast constraint satisfaction

Reproducibility: baselines + QA artifacts are archived on Zenodo. Code lives in public GitHub repos with tests and docs.


🧪 Reproducible records (Zenodo)

Community Research Engagement