fokker-planck-analyzer

Name: fokker-planck-analyzer
Availability: InStock
Author: plurigrid

Compare empirical weight distributions to Gibbs distributions for Langevin training convergence.

60|13|Updated Dec 22, 2025

Search Tags:#gibbs-distribution #fokker-planck #langevin-dynamics #mixing-time #kl-divergence

Ready to run

npx skills add https://github.com/plurigrid/asi --skill fokker-planck-analyzer

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: fokker-planck-analyzer
Source: https://github.com/plurigrid/asi/tree/main/skills/fokker-planck-analyzer
Command: npx skills add https://github.com/plurigrid/asi --skill fokker-planck-analyzer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Layer 5: Convergence to Equilibrium Analysis; verifies that training under Langevin dynamics reaches Gibbs equilibrium as predicted by Fokker-Planck theory.

Core Features & Use Cases

Gibbs convergence checks: compare empirical distribution to theoretical Gibbs distribution.
Mixing-time estimation: approximate time to equilibrium from Hessian spectrum or empirical convergence.
Steady-state validation: KL divergence and gradient stability checks.

Quick Start

just langevin-solve net=network T=0.01 n_steps=1000 just fokker-check-convergence

Frequently Asked Questions about fokker-planck-analyzer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I verify that neural network training reaches Gibbs equilibrium under Langevin dynamics?▼

Gibbs equilibrium verification compares your empirical weight distribution to the theoretical Gibbs distribution predicted by Fokker-Planck theory. This Skill analyzes training trajectories across loss landscapes and temperatures to confirm convergence to equilibrium, supporting mixing-time estimation and steady-state validation.

What is mixing time in Langevin dynamics and how do I estimate it?▼

Mixing time is the duration required for a system to reach equilibrium under Langevin dynamics. This Skill approximates mixing time from Hessian eigenvalues or empirical convergence data, enabling you to predict how long training needs to stabilize at a given temperature.

How do I track KL divergence convergence during neural network training?▼

KL divergence tracking measures how quickly your empirical weight distribution approaches the Gibbs distribution. This Skill monitors KL divergence over training epochs and validates gradient stability, reporting convergence metrics and steady-state assessments.

Can I use Fokker-Planck analysis to validate training across different loss landscapes and temperatures?▼

Yes. This Skill applies Fokker-Planck-based convergence checks across various loss landscapes and temperature settings. It computes Gibbs convergence metrics and validates steady-state behavior, supporting scenarios from low-temperature steep landscapes to high-temperature exploration regimes.

What's the relationship between Hessian eigenvalues and equilibrium convergence in Langevin dynamics?▼

Hessian eigenvalues determine the local curvature of your loss landscape and directly influence mixing time under Langevin dynamics. This Skill uses the Hessian spectrum to estimate equilibrium convergence speed, helping you understand how landscape geometry affects training dynamics.