👋 Hello, AstanaHub Community! I am pleased to announce the release of my project `OIKAN v0.0.3`, a neuro-symbolic machine learning framework for creating interpretable models, which is being developed as part of my thesis project.

> OIKAN (v0.0.3) is a neuro-symbolic machine learning framework inspired by Kolmogorov-Arnold Representation Theory (KART), designed to deliver accurate, interpretable, and efficient models for tabular data. By integrating the strengths of neural networks and symbolic regression, OIKAN produces human-readable mathematical formulas that approximate complex data relationships while maintaining high predictive performance. Unlike traditional approaches, which suffer from high computational complexity, OIKAN employs a streamlined approach that balances interpretability, efficiency, and accuracy.

Key Features:

🧠 Neuro-Symbolic ML: Combines neural network learning with symbolic mathematics;

🔢 Automatic Formula Extraction: Generates human-readable mathematical expressions;

🎯Scikit-learn Compatible: Familiar .fit() and .predict() interface; 

🔬 Research-Focused: Designed for academic exploration and experimentation;

📊 Multi-Task: Supports both regression and classification problems;  

🔗 Links to the project:

> GitHub: https://github.com/silvermete0r/oikan 

> PyPI: https://pypi.org/project/oikan/ 

> DeepWiki: https://deepwiki.com/silvermete0r/oikan

> Get Started Template Notebook: https://www.kaggle.com/code/armanzhalgasbayev/oikan-v0-0-3-get-started-template-notebook


🙌 How can you help in the development of the project?

> Try to use OIKAN;

> Leave a feedback in comments or here (in GitHub Discussions): https://github.com/silvermete0r/oikan/discussions/50

> Answer the poll about usability (in GitHub Discussions): https://github.com/silvermete0r/oikan/discussions/49

OIKAN v0.0.3 - High-level Architecture:

#OIKANv003 #NeuroSymbolicML #InterpretableML