🔴 ENTRO-CORE — Regime-Dependent Entropy-Augmented Control
"Intelligence by Design, Stability by Physics"
ENTRO-CORE is the third project of the EntropyLab research program (E-LAB-03). It investigates entropy-based control architectures for dynamical systems, introducing a hybrid regime-switching controller that combines PID with nonlinear entropy feedback.
Project Code: E-LAB-03 | Lab: Entropy Research Lab | Submitted: April 2026
📋 Table of Contents
- Overview
- Key Finding
- Control Laws
- Experimental Results
- Project Structure
- Installation
- Quick Start
- Validation
- EntropyLab Roadmap
- Documentation
- Citation
- Author
- License
🔭 Overview
Modern AI systems operate under high-dimensional feedback loops that can lead to instability, especially under long-context reasoning or recursive inference. ENTRO-CORE moves entropy control from an external monitoring layer (ENTRO-AI) into the core control architecture itself.
Instead of: Monitor → Detect → Intervene (reactive)
ENTRO-CORE does: Self-regulate internally (proactive)
| Metric | Value |
|---|---|
| PID Final Ψ (stable regime) | 0.017 |
| ENTRO-CORE Final Ψ | -0.239 |
| Hybrid Final Ψ | -0.012 |
| Validation Runs | 5 runs per controller |
| Noise Robustness | σ = 0.00 → 0.15 |
🔬 Key Finding
Entropy-based control is regime-dependent, not universally superior.
PID remains optimal in stable linear systems, while entropy-based control provides benefits in near-critical regimes.
| Regime | Optimal Controller |
|---|---|
| Stable linear | PID |
| Near-critical / nonlinear | ENTRO-CORE / Hybrid |
📐 Control Laws
PID Controller
u_PID(t) = K_p·e(t) + K_i·∫e(τ)dτ + K_d·ė(t)
with e(t) = -Ψ(t), K_p=0.8, K_i=0.2, K_d=0.3
Entropy-Based Controller (ENTRO-CORE v1)
u_ENTRO(t) = w₁·σ(Ψ_norm - θ) + w₂·tanh(Ψ̇) + w₃·tanh(Ψ̈)
with w₁=0.5, w₂=0.3, w₃=0.2, θ=1.4
Hybrid Regime-Switching Controller
u(t) = { u_PID(t) if Ψ(t) < Ψ_th
{ u_ENTRO(t) if Ψ(t) ≥ Ψ_th
with switching threshold Ψ_th = 1.7
📊 Experimental Results
Under near-critical initial conditions (Ψ(0)=1.8, Ψ̇(0)=0.3):
| Controller | Final Ψ (t=20s) | Outcome |
|---|---|---|
| Uncontrolled | 0.053 | Naturally stable |
| PID | 0.017 | Optimal convergence |
| ENTRO-CORE v1 | -0.239 | Mild overshoot |
| Hybrid (threshold=1.7) | -0.012 | Robust performance |
Noise Sensitivity (Hybrid Controller)
| Noise σ | Final Ψ | Outcome |
|---|---|---|
| 0.00 | 0.339 | Stabilized |
| 0.02 | 0.341 | Stabilized |
| 0.05 | 0.355 | Stabilized |
| 0.10 | 0.412 | Stabilized |
| 0.15 | 0.523 | Stabilized |
🗂️ Project Structure
entro-core/
│
├── 📄 README.md # This file
├── 📄 LICENSE # MIT License
├── 📄 CHANGELOG.md # Version history
├── 📄 CONTRIBUTING.md # Contribution guidelines
├── 📄 CITATION.cff # Academic citation metadata
├── 📄 pyproject.toml # Build configuration
├── 📄 requirements.txt # Runtime dependencies
├── 📄 requirements-dev.txt # Development dependencies
│
├── 📁 entro_core/ # Core Python package
│ ├── 📄 init.py # Package entry point
│ ├── 📄 controller.py # ENTRO-CORE v1 (original)
│ ├── 📄 controller_v2.py # ENTRO-CORE v2 (negative feedback)
│ ├── 📄 hybrid_controller.py # Regime-switching controller
│ ├── 📄 actuator.py # Actuation strategies
│ ├── 📄 normalize.py # Logistic normalization (Eq. 4)
│ ├── 📄 state.py # State tracker (Ψ, Ψ̇, Ψ̈)
│ └── 📄 control_law.py # Control law u(t) (Eq. 5)
│
├── 📁 simulation/ # Validation simulations
│ ├── 📄 generate_figures_fixed.py # Trajectory generation
│ ├── 📄 validation_experiment.py # Full validation suite
│ ├── 📄 hybrid_test.py # Hybrid controller test
│ └── 📁 results/
│ └── 📄 trajectories_fixed.csv # Comparison data
│
├── 📁 tests/ # Unit tests
│ ├── 📄 test_controller.py # Controller tests (21 tests)
│ └── 📄 init.py
│
├── 📁 paper/ # Research paper assets
│ └── 📁 arxiv/
│ └── 📄 entro_core_paper_revised.tex # LaTeX manuscript
│
└── 📁 docs/ # Documentation
├── 📄 index.md
├── 📄 control_theory.md
└── 📄 api_reference.md
⚙️ Installation
Requirements
- Python 3.11+
- No external dependencies required (pure Python)
Via PyPI
pip install entro-core
From Source
git clone https://github.com/gitdeeper10/entro-core.git
cd entro-core
pip install -e ".[dev]"
🚀 Quick Start
- Import and Use Hybrid Controller
from entro_core.hybrid_controller import HybridController
# Create controller with threshold
controller = HybridController(threshold=1.7)
# Simulate a near-critical state
result = controller.step(psi=1.8)
print(f"u(t) = {result.u:.3f}")
print(f"Mode: {result.mode}") # 'PID' or 'ENTRO-CORE'
print(f"Active: {result.is_active}")
u(t) = 0.423
Mode: ENTRO-CORE
Active: True
- Run Validation Experiment
python simulation/validation_experiment.py
- Generate Comparison Trajectories
python simulation/generate_figures_fixed.py
📊 Validation Results
Test Suite (21 tests)
Category Passed Normalization 4/4 ✅ Sigmoid 3/3 ✅ Tanh 3/3 ✅ Control Law 3/3 ✅ Actuation 4/4 ✅ Controller 4/4 ✅ Total 21/21
🗺️ EntropyLab Research Roadmap
E-LAB-01 ✅ ENTROPIA — Thermodynamic unification
E-LAB-02 ✅ ENTRO-AI — Entropy-resistant AI inference
E-LAB-03 ✅ ENTRO-CORE — Regime-dependent control (this repository)
E-LAB-04 📅 ENTRO-ENGINE — Generalized entropy control framework
E-LAB-05 📅 ENTRO-FIN — Financial entropic dynamics
E-LAB-06 📅 ENTRO-SOCIAL — Information cascades in networks
E-LAB-07 📅 ENTRO-QUANTUM — Quantum entropy extension
E-LAB-08 📅 ENTRO-BIO — Biological neural entropy systems
E-LAB-09 📅 ENTRO-CLIMATE — Climate entropy modeling
E-LAB-10 📅 ENTRO-META — Unified entropy control theory
✅ Complete | 🔄 In Progress | 📅 Planned
📚 Documentation
Resource Link Full Documentation entropia-lab.netlify.app/entro-core/docs Research Paper (PDF) entropia-lab.netlify.app/entro-core/paper API Reference entropia-lab.netlify.app/entro-core/api Parent Project (ENTRO-AI) entro-ai.netlify.app Foundation (ENTROPIA) doi.org/10.5281/zenodo.19416737
🤝 Contributing
git clone https://github.com/gitdeeper10/entro-core.git
cd entro-core
pip install -e ".[dev]"
pytest tests/
Priority contribution areas:
· Smooth switching (sigmoidal interpolation) · Additional system dynamics models · Real-world validation datasets
📖 Citation
@software{baladi2026entrocore,
author = {Samir Baladi},
title = {ENTRO-CORE: Regime-Dependent Entropy-Augmented Control},
year = {2026},
version = {0.1.0},
publisher = {Zenodo},
doi = {10.5281/zenodo.19431029},
url = {https://doi.org/10.5281/zenodo.19431029},
note = {E-LAB-03. Builds on E-LAB-01 and E-LAB-02}
}
Parent frameworks:
@article{baladi2026entropia,
title = {ENTROPIA: Statistical Dynamics of Information Dissipation},
author = {Samir Baladi},
year = {2026},
doi = {10.5281/zenodo.19416737},
note = {E-LAB-01}
}
@software{baladi2026entroai,
author = {Samir Baladi},
title = {ENTRO-AI: Entropy-Resistant Inference Architecture},
year = {2026},
version = {2.0.0},
doi = {10.5281/zenodo.19284086},
note = {E-LAB-02}
}
👤 Author
Samir Baladi
· Role: Principal Investigator, Interdisciplinary AI Researcher · Affiliation: Ronin Institute / Rite of Renaissance · Email: gitdeeper@gmail.com · ORCID: 0009-0003-8903-0029 · GitHub: github.com/gitdeeper10 · GitLab: gitlab.com/gitdeeper10
📜 License
MIT License — see LICENSE for details.
ENTRO-CORE — Entropy Research Lab — E-LAB-03
Regime-Dependent Entropy-Augmented Control
https://img.shields.io/badge/PyPI-pip_install_entro--core-red.svg https://img.shields.io/badge/DOI-10.5281/zenodo.19431029-blue.svg
entropia-lab.netlify.app · pip install entro-core
Builds on ENTROPIA (E-LAB-01) and ENTRO-AI (E-LAB-02)
Metadata
Release files for entro-core 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| entro_core-0.1.0.tar.gz | 16.9 kB | Without provenance | Details for entro_core-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| entro_core-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for entro_core-0.1.0-py3-none-any.whl |
Total release size: 33.0 kB
Release files / entro_core-0.1.0.tar.gz
| Download URL | entro_core-0.1.0.tar.gz |
|---|---|
| Size | 16.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
59212c76a18876e1b2c8484f0e83bb982510d30245eb3a2f8bac5692b686e79a
|
|
BLAKE2b-256 checksum How to use checksums |
6dd63916fc4b4b327647e252e80c393ffd6b63d25e671a80b5ea2a2fcb497abf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
ENTRO-CORE-Uploader/0.1.0
|
Release files / entro_core-0.1.0-py3-none-any.whl
| Download URL | entro_core-0.1.0-py3-none-any.whl |
|---|---|
| Size | 16.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
202fb1bf43148eb7b5f62afbd32315b643b25cb83965b4c9a52706ed8397eb9c
|
|
BLAKE2b-256 checksum How to use checksums |
75920df2f13b9082e76450521291e621e367f2577569a200060b6878b496e9c0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
ENTRO-CORE-Uploader/0.1.0
|
Security for entro-core 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.