8-component game-theoretic field mapping equity × pot-odds space. Monte Carlo simulation feeds a continuous decision surface — play against it live.
Open portfolio/index.html in your browser to:
- Play Texas Hold'em against the AI with real-time field analysis
- Explore the Model — formula breakdown, architecture pipeline, game theory foundations
- Watch the heatmap shift as each card is dealt and every bet is placed
The engine evaluates every decision through an 8-component continuous field:
Ψ(e, p) = V_equity + R_raise + C_call + F_fold + B_bluff + S_street + J_jump + Γ_strength
| Component | Force | Description |
|---|---|---|
| V | Equity Peak | Gaussian centered on hero's equity position |
| R | Raise Pressure | Aggression surface with street multiplier |
| C | Call Defense | Defensive ridge at ~45% equity |
| F | Fold Gravity | Negative basin pulling weak hands to fold |
| B | Bluff Surface | Deception peaks for profitable bluff spots |
| J | Jump Threshold | Sharp sigmoid modeling decision boundaries |
Monte Carlo (2,400 rollouts) → ζ-Field Eval (8 components) → Decision (RAISE/CALL/FOLD)
broki/
├── portfolio/ # Web application
│ ├── index.html # Main page — hero, model deep-dive, poker game, about
│ ├── style.css # Full dark-theme styling with glassmorphism
│ ├── main.js # Hero canvas animation, scroll effects, nav
│ ├── poker/
│ │ ├── engine.js # Card evaluation, Monte Carlo simulation, hand ranking
│ │ ├── field.js # ζ-field computation, commentary, heatmap rendering
│ │ ├── game.js # Game state machine, AI opponent logic
│ │ └── ui.js # DOM rendering, user controls, structured commentary
│ └── assets/ # GIFs and images
├── poker_game_field.py # Python implementation of the ζ-field model
├── zeta_field_advanced.py # Advanced field with Heston/GARCH/Lévy components
├── zeta_field_python.py # Core ζ-field Python implementation
├── zeta_field_r.R # R implementation
├── ZetaField.java # Java implementation
└── kuhn_poker_cfr.py # Counterfactual Regret Minimization for Kuhn Poker
# Clone the repo
git clone https://github.com/Zhandolia/zeta-engine.git
cd zeta-engine
# Serve the portfolio locally
cd portfolio
python3 -m http.server 8080
# Open http://localhost:8080- Real-time ζ-field heatmap — watch the action surface shift with every card
- Model Insight panel — structured commentary with headline, stats, and context
- Live Analysis sidebar — equity, pot odds, EV, confidence per hand
- AI opponent — makes decisions using the same field model
- Suggestion bar — real-time recommended action with confidence
| File | Language | Description |
|---|---|---|
poker_game_field.py |
Python | Full poker field model with visualization |
zeta_field_advanced.py |
Python | Heston SV, GARCH, Merton jump-diffusion |
ZetaField.java |
Java | Java port of the ζ-field model |
zeta_field_r.R |
R | R implementation for statistical analysis |
kuhn_poker_cfr.py |
Python | CFR algorithm on simplified Kuhn Poker |
MIT