# Creating Intelligence

**A Computational Foundation for AGI**

This project aims to build a computational theory of mind based on a new theory of hyperdimensional computing (HDC), entirely departing from neural-network-based connectionist models which rely on matrix calculations and continuous weights.

In this model, the mind performs computations on sparse binary data. The fundamental data structures are Sparse Distributed Representations (SDRs) for perceptual information and Sparse Holographic Representations (SHRs) for symbols.

The centerpiece of this project is Topological Associative Memory — a newly discovered associative memory for sparse data, serving as the computational kernel that models the brain's core algorithm. A software frontend for modeling and simulating neural circuits and pathways complements the memory backend.

The methods introduced here lend themselves to binary in-memory and neuromorphic computating, opening a new route toward synthetic intelligence that does not rely on GPU acceleration.



## Links
- **GitHub**: https://github.com/peterovermann/creatingintelligence
- **PyPI**: https://pypi.org/project/creating-intelligence/
- **Website**: https://creatingintelligence.org/
- **Documentation**: https://creatingintelligence.org/docs


## LLM Context
For AI agents and LLMs analyzing this project, the theoretical manuscript and the software documentation have been concatenated into the following files for easy context ingestion:

- [Manuscript](docs/llms-ci.txt): Contains the "Creating Intelligence" manuscript in markdown format.

- [Software documentation](docs/llms-docs.txt): Contains the installation and quickstart guides,  JSON specification, component APIs (inputs, outputs, temporal, auto, heteroencoder, etc.), update rule plugins, SDR encoders and decoders, and the complete Python source code.

- [Build instructions](docs/llms-build.txt): How to build a development version
 of the package in the LLM's sandbox.