[Article.Ai] Supersymmetry used on basic perceptron by Agoristas, … Biroli et al.

Supersymmetry was recently used to find the dynamical mean field equations (DMFE) that describe the Langevin dynamics for the random continuous perceptron model, by Biroli et al (2018) — (They replace Real-valued Perceptron Hamiltonian-based neuron states with SUSY fields, to describe Langevin dynamics of the basic Perceptron, under a probability distribution of Real valued weights. I.e. Real Valued Weights are learned.):

https://arxiv.org/abs/1710.04894

The above may potentially be used as motivation; to further advance my “Thought Curvature” or “Supersymmetric Artificial Neural Network (SANN)” idea (SANN concerns the Super-Hamiltonian, under a probability distribution of Supersymmetric weights. I.e. Supersymmetric Weights are learned.)

My machine learning model: https://github.com/JordanMicahBennett/Supersymmetric-artificial-neural-network

SuperHamiltonian physics work by another author: https://arxiv.org/abs/hep-th/0506170

▼

▼

Author:

I am an atheist, casual body builder, and software engineer