Update README.md

# Dualist: Hybrid Othello AI

Dualist is a high-performance Othello agent utilizing a hybrid architecture that integrates **PyTorch** with the world-class engine **Edax**

## Architecture
The system is built around a triad of interacting components[cite: 10]:
**The Student (PyTorch):** A deep neural network (ResNet) featuring a dual-head structure for Policy and Value prediction.
**The Teacher (Edax):** Functions as an "Oracle" by providing ground-truth evaluations via an optimized C-based bitboard engine.
**Experience Replay Buffer:** Stores millions of positions in LMDB format to break correlation and stabilize training.

## Technical Specifications
**Input:** A (3, 8, 8) tensor encoding the current player's pieces, the opponent's pieces, and the current turn.
**Training Methodology:** A Teacher-Student Curriculum transitioning from Supervised Bootstrapping to Reinforcement Learning with dynamic search depth.
**Integration:** High-performance Python bridge via `ctypes` to call Edax functions directly in memory without CLI overhead.

## Deployment & Usage
The model is designed to operate within a modern stack including:
* **FastAPI** for the inference API.
* **PostgreSQL** for match history and analytical storage.
* **Vite / React Native** for cross-platform frontend interaction.



README.md

@@ -1,3 +1,14 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+language:
+- en
+- sv
+metrics:
+- accuracy
+tags:
+- othello
+- reinforcement-learning
+- alphazero
+- edax
+- board-games
+---