PyHGF: A neural network library for predictive coding

Deep active inference agents using Monte-Carlo methods

PyTorch library for Active Fine-Tuning

Official Implementation for the paper "SR-AIF: Solving Sparse-Reward Robotic Tasks from Pixels with Active Inference and World Models"

A sovereign cognitive architecture with IIT 4.0 integrated information, residual-stream affective steering (CAA), Global Workspace Theory, active inference, and 72 consciousness modules — running locally on Apple Silicon.

Implementation/simulation of active neural generative coding (ANGC) for training neurobiologically-plausible active inference agent models.

Manuscript source: Self-orthogonalizing attractor neural networks emerging from the free energy principle

MLSS-2026 Melbourne Bert's lectures

[NeurIPS 2021] World modelling and action learning using a contrastive formulation of the active inference framework, for reaching visual goal states

Cognitive science research workspace — Active Inference, Bayesian modeling, ant behavior, and computational neuroscience

PID-like control implemented as active inference with linear generative models

Active Inference & Category Theory

Homing Piegon is an inference framework implementing Variational Message Passing. It can be used to implement an Active Inference agent that performs planning using a Tree Search algorithm that can been seen as a form of Bayesian Model Expansion.

Active inference agent and corresponding environment in Unity used in the study "A deep active inference model of the rubber-hand illusion"

Deep Active Inference (Deep AIF) Agents

DIE — is an Artificial Life project aimed at reproducing emergence of distributed intelligence under environmental pressures using learning cellular automata models.