Free-energy minimization in joint agent-environment systems: A niche construction perspective

JOURNAL OF THEORETICAL BIOLOGY

卷:455

Article

Bruineberg, Jelle^1,2  Rietveld, Erik^1,2,4,6  Parr, Thomas³  van Maanen, Leendert^2,5  Friston, Karl J.³ 

[1] Univ Amsterdam, Dept Philosophy, Inst Log Language & Computat, Amsterdam, Netherlands

[2] Univ Amsterdam, Amsterdam Brain & Cognit Ctr, Amsterdam, Netherlands

[3] UCL, Wellcome Trust Ctr Neuroimaging, Inst Neurol, London WC1N 3BG, England

[4] Univ Amsterdam, Acad Med Ctr, Dept Psychiat, Amsterdam, Netherlands

[5] Univ Amsterdam, Dept Psychol, Amsterdam, Netherlands

[6] Univ Twente, Dept Philosophy, Enschede, Netherlands

关键词: Active inference;    Free energy principle;    Markov decision processes;    Niche construction;    Agent-environment complementarity;    Adaptive environments;    Desire paths;   

DOI  :  10.1016/j.jtbi.2018.07.002

来源: Elsevier

【 摘 要 】

The free-energy principle is an attempt to explain the structure of the agent and its brain, starting from the fact that an agent exists (Friston and Stephan, 2007; Friston et al., 2010). More specifically, it can be regarded as a systematic attempt to understand the 'fit' between an embodied agent and its niche, where the quantity of free-energy is a measure for the 'misfit' or disattunement (Bruineberg and Rietveld, 2014) between agent and environment. This paper offers a proof-of-principle simulation of niche construction under the free-energy principle. Agent-centered treatments have so far failed to address situations where environments change alongside agents, often due to the action of agents themselves. The key point of this paper is that the minimum of free-energy is not at a point in which the agent is maximally adapted to the statistics of a static environment, but can better be conceptualized an attracting manifold within the joint agent-environment state-space as a whole, which the system tends toward through mutual interaction. We will provide a general introduction to active inference and the free-energy principle. Using Markov Decision Processes (MDP5), we then describe a canonical generative model and the ensuing update equations that minimize free-energy. We then apply these equations to simulations of foraging in an environment; in which an agent learns the most efficient path to a pre-specified location. In some of those simulations, unbeknownst to the agent, the 'desire paths' emerge as a function of the activity of the agent (i.e. niche construction occurs). We will show how, depending on the relative inertia of the environment and agent, the joint agent-environment system moves to different attracting sets of jointly minimized free-energy. (C) 2018 The Author(s). Published by Elsevier Ltd.

【 授权许可】

Free