hebbian-learning

Raise a neural network as a pet - a transparent cognitive sandbox where a digital squid learns, grows, and rewires itself via Hebbian learning & Neurogenesis

NGC-Learn: Computational Neuroscience and NeuroAI in Python

Meta-Learning through Hebbian Plasticity in Random Networks: https://arxiv.org/abs/2007.02686

A lightweight and flexible framework for Hebbian learning in PyTorch.

Hopfield network implemented with Python

Python implementation of the Epigenetic Robotic Architecture (ERA). It includes standalone classes for Self-Organizing Maps (SOM) and Hebbian Networks.

PyPi Package of Self-Organizing Recurrent Neural Networks (SORN) and Neuro-robotics using OpenAI Gym

This repository has implementations of various alternatives to backpropagation for training neural networks.

Persistent memory for AI coding agents. Your agent learns your codebase the way a senior engineer would — what files go together, what you usually edit next. Works with Claude Code, Cursor, Cline, Continue. 100% local. (Side effect: 40-70× cheaper code questions.)

Code and paper reproductions for MLPs Are Hebbians

Biologically inspired language model using Jaccard Surprise as its only training signal. No backprop. No GPU. Online Hebbian learning from corrections. Two-layer cortex with apical feedback. Runs on CPU under 200MB.

Code for paper NeurIPS AMHN 2023

Code for Limbacher, T. and Legenstein, R. (2020). H-Mem: Harnessing synaptic plasticity with Hebbian Memory Networks

A Computational Substrate for Self-Organizing Biologically-Plausible AI

Code for Limbacher, T., Özdenizci, O., & Legenstein, R. (2022). Memory-enriched computation and learning in spiking neural networks through Hebbian plasticity. arXiv preprint arXiv:2205.11276.

Contrastive Hebbian learning on MNIST, reaching ~97% accuracy with a small 784-128-10 MLP.

Malleable spiking neural network framework and training platform.

A neural network model builder, leveraging a neuro-symbolic interface.

Non-LLM AI substrate that remembers who it is across sleep gaps. Pure numpy, Hebbian + sleep replay + persistent slow weights. Ten identity tests passing in toy worlds. Research artifact.

In this project, I used Hebbian, Perceptron and Adaline neural networks to implement AND gate, and OR gate.