Generative models, linguistic communication and active inference

NEUROSCIENCE AND BIOBEHAVIORAL REVIEWS

卷:118

Article

Friston, Karl J.¹  Parr, Thomas¹  Yufik, Yan²  Sajid, Noor¹  Price, Catherine J.¹  Holmes, Emma¹ 

[1] Wellcome Ctr Human Neuroimaging, Inst Neurol, UCL Queen Sq,12 Queen Sq, London WC1N 3AR, England

[2] Virtual Struct Res Inc, 12204 St James Rd, Potomac, MD 20854 USA

关键词: Message passing;    Inference;    Connectivity;    Free energy;    Neuronal;    Hierarchical;    Bayesian;    Language;   

DOI  :  10.1016/j.neubiorev.2020.07.005

来源: Elsevier

【 摘 要 】

This paper presents a biologically plausible generative model and inference scheme that is capable of simulating communication between synthetic subjects who talk to each other. Building on active inference formulations of dyadic interactions, we simulate linguistic exchange to explore generative models that support dialogues. These models employ high-order interactions among abstract (discrete) states in deep (hierarchical) models. The sequential nature of language processing mandates generative models with a particular factorial structure-necessary to accommodate the rich combinatorics of language. We illustrate linguistic communication by simulating a synthetic subject who can play the 'Twenty Questions' game. In this game, synthetic subjects take the role of the questioner or answerer, using the same generative model. This simulation setup is used to illustrate some key architectural points and demonstrate that many behavioural and neurophysiological correlates of linguistic communication emerge under variational (marginal) message passing, given the right kind of generative model. For example, we show that theta-gamma coupling is an emergent property of belief updating, when listening to another.

【 授权许可】

Free