Bayesian-based Markovian-model

BbMm, aims to become a multi-language API to create deliberative agent based on decision-making under uncertainty. The approach relies on Markov Decision Process with transition based on Bayesian Network model.

Actually, BbMm include 2 main components:

• libBbMm : The core C librairie of BbMm [on github].
• wraps-python : The python wraps of the libBbMm modules [on github].

Concurency:

BbMm is not what you looking for ? Look at those solutions:

• SaLinA : https://arxiv.org/pdf/2110.07910.pdf - https://github.com/facebookresearch/salina
• MDPToolBox :
• LibDAI : C++ lib for Discrete Approximate Inference
• gobnilp : Bayesian Network learning using Integer Linear Programming
• torchhd : Python lib for Hyperdimensional Computing (or Vector Symbolic Architectures) Artticle: https://jmlr.org/papers/v24/23-0300.html
• farama : à fouiller.