Wassim Tenachi

GitHub - WassimTenachi/PhySO: Physical Symbolic Optimization Physical Symbolic Optimization. Contribute to WassimTenachi/PhySO development by creating an account on GitHub.

✨ Ever wanted to try Symbolic Regression?
Φ-SO just got simpler!

📦 One-line install
🐍 Works seamlessly with:
- Python 3.12
- Latest NumPy/PyTorch/SymPy

No dependency hell — just clean equation discovery.

Get started:
github.com/WassimTenachi/PhySO

GitHub - WassimTenachi/PhySO: Physical Symbolic Optimization Physical Symbolic Optimization. Contribute to WassimTenachi/PhySO development by creating an account on GitHub.

More than curve-fitting, Φ-SO searches in the space of mathematical expressions themselves - guided by a neural policy and physics (e.g. units, symmetries).

Follow for more about automated equation discovery ! :)

code: github.com/WassimTenachi/PhySO
paper: arxiv.org/abs/2303.03192

The video shows Φ-SO - a Physical Symbolic Optimization algorithm - rediscovering the tariff law implemented in the US — directly from data.
No prior knowledge. Just learning patterns.

It uses deep RL to discover interpretable symbolic equations that make sense to humans by trial-and-error.

Retro-engineering the US tariffs formula with deep reinforcement learning 👀