ago109/active-neural-generative-coding

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

30
/ 100
Emerging

This project helps researchers and cognitive scientists simulate and study how a neurobiologically-plausible agent learns to solve control problems without needing dense rewards. It takes in basic experimental parameters and outputs numerical arrays of episodic rewards, epistemic signals, and agent positions, along with visual plots of these metrics. This is for computational neuroscientists or AI researchers exploring alternative learning mechanisms in agents.

No commits in the last 6 months.

Use this if you are a researcher interested in simulating biologically-plausible, backpropagation-free learning for active inference agents in simple control environments with sparse rewards.

Not ideal if you are looking for a robust, production-ready reinforcement learning algorithm for complex, real-world engineering problems.

computational-neuroscience cognitive-modeling active-inference reinforcement-learning-research biologically-inspired-ai
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

22

Forks

2

Language

Python

License

BSD-3-Clause

Last pushed

Feb 19, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ago109/active-neural-generative-coding"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.