arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2306.05053cs.NEcs.AI
收藏

Hebbian学习网络中的主动推断

Active Inference in Hebbian Learning Networks

imec(比利时微电子研究中心);KU Leuven(荷语鲁汶大学);University of California at San Diego(加州大学圣地亚哥分校);VERSES Research Lab(VERSES研究实验室)

Ali Safa, Tim Verbelen, Lars Keuninckx, Ilja Ocket, André Bourdoux, Francky Catthoor, Georges Gielen, Gert Cauwenberghs

AI总结:

本研究提出基于Hebbian可塑性网络的主动推断方法,在Mountain Car环境中优于Q-learning,且无需重放缓冲区,为设计能学习环境动态的AIF网络提供新思路。

摘要:

本研究探讨了配备局部Hebbian可塑性的脑启发神经集成如何执行主动推断(AIF)以控制动态智能体。一个捕捉环境动态的生成模型由包含两个不同Hebbian集成的网络学习:后验网络,根据观测推断潜在状态;以及状态转移网络,根据当前状态-动作对预测下一个期望的潜在状态。实验研究使用OpenAI gym套件中的Mountain Car环境进行,以研究各种Hebbian网络参数对任务性能的影响。结果表明,所提出的Hebbian AIF方法优于使用Q-learning,且不需要像典型强化学习系统那样的重放缓冲区。这些结果激励了对Hebbian学习的进一步研究,以设计能够学习环境动态而无需重访过去缓冲经验的AIF网络。

英文摘要:

This work studies how brain-inspired neural ensembles equipped with local Hebbian plasticity can perform active inference (AIF) in order to control dynamical agents. A generative model capturing the environment dynamics is learned by a network composed of two distinct Hebbian ensembles: a posterior network, which infers latent states given the observations, and a state transition network, which predicts the next expected latent state given current state-action pairs. Experimental studies are conducted using the Mountain Car environment from the OpenAI gym suite, to study the effect of the various Hebbian network parameters on the task performance. It is shown that the proposed Hebbian AIF approach outperforms the use of Q-learning, while not requiring any replay buffer, as in typical reinforcement learning systems. These results motivate further investigations of Hebbian learning for the design of AIF networks that can learn environment dynamics without the need for revisiting past buffered experiences.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑