Welcome to the Mieza documentation. The guides and concepts here will teach you how to build systems that can make **well-grounded strategic** decisions.

The core idea behind this is **Optimal Strategic Reasoning (OSR)**, which is the application of game theory principles to automated decision-making. In well-defined strategic situations, it can help systems make choices that are mathematically sound.

Unlike traditional approaches that optimize for average performance, OSR focuses on worst-case guarantees and mathematical rigor. This provides a foundation for making decisions when interacting with other systems or people in competitive or cooperative settings.

## Core Properties of OSR

OSR provides several properties that are critical for building trustworthy and effective systems, especially in high-stakes environments.

### Robustness Against Adversaries
Traditional machine learning often fails against adversarial inputs. For well-defined games, OSR provides robustness by assuming that an opponent will actively try to exploit any weakness. The resulting strategies are **unexploitable**: in two-player zero-sum settings, no opponent can do better than break even against an equilibrium strategy. This property holds at the solver level, where the equilibrium is computed.

### Provability and Verification
OSR models are built on mathematical foundations. For well-defined games, the algorithms (such as CFR) are known to converge to Nash equilibrium, and the resulting strategies can be rigorously analyzed. The guarantees are strongest for games with clear rules, known payoffs, and finite structure.

### Interpretability
OSR decisions are inherently interpretable. Because the reasoning is based on explicit game trees and utility calculations, it's possible to **trace the logic** behind any decision and **explain why** a particular action was chosen over alternatives.

By going through these docs, you'll learn:
- The core **concepts** behind strategic reasoning.
- The **tutorials** for building your first automated system.
- The **API references** for integrating OSR into your applications.

Let's begin!