Toward Design of Synthetic Active Inference Agents by Mere Mortals

TL;DR

This paper discusses the development of a software toolbox designed to make creating active inference agents accessible to non-experts, addressing computational challenges on edge devices and promoting democratization of this technology.

Contribution

It introduces a new toolbox aimed at simplifying the design of active inference agents for non-expert users, facilitating their deployment on resource-limited hardware.

Findings

Conceptual framework for a user-friendly toolbox

Initial implementation supporting active inference on edge devices

Potential to democratize active inference agent development

Abstract

The theoretical properties of active inference agents are impressive, but how do we realize effective agents in working hardware and software on edge devices? This is an interesting problem because the computational load for policy exploration explodes exponentially, while the computational resources are very limited for edge devices. In this paper, we discuss the necessary features for a software toolbox that supports a competent non-expert engineer to develop working active inference agents. We introduce a toolbox-in-progress that aims to accelerate the democratization of active inference agents in a similar way as TensorFlow propelled applications of deep learning technology.