arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.01677cs.LGcs.NEq-bio.NC
收藏

一种元学习的对比规则

A contrastive rule for meta-learning

ETH Zürich(苏黎世联邦理工学院);University of Zürich(苏黎世大学)

Nicolas Zucchet, Simon Schug, Johannes von Oswald, Dominic Zhao, João Sacramento

摘要:

人类和其他动物能够在解决给定问题领域的相关任务时提高学习性能,从而能够从极其有限的数据中学习。虽然突触可塑性通常被认为是大脑学习的基础,但学习过程通过经验改善的精确神经和突触机制还没有得到很好的理解。在这里,我们提出了一个通用的,生物学上合理的元学习规则,估计梯度相对于一个基本的学习算法的参数,只需运行两次。我们的规则可以理解为对比赫布学习到元学习的推广,值得注意的是,它既不需要计算二阶导数,也不需要时间回溯,这是以前基于梯度的方法的两个特征,在物理神经回路中很难想象。我们证明了我们的规则的一般性,通过将其应用到两个不同的模型:一个复杂的突触与内部状态,巩固任务共享的信息,和一个双系统架构,其中一个主要的网络是快速调制的另一个学习每个任务的细节。对于这两种模型,我们的元学习规则在广泛的基准问题上匹配或优于参考算法,同时仅使用假定在神经元和突触处局部可用的信息。我们证实了这些发现与理论分析的梯度估计误差所产生的规则。

英文摘要:

Humans and other animals are capable of improving their learning performance as they solve related tasks from a given problem domain, to the point of being able to learn from extremely limited data. While synaptic plasticity is generically thought to underlie learning in the brain, the precise neural and synaptic mechanisms by which learning processes improve through experience are not well understood. Here, we present a general-purpose, biologically-plausible meta-learning rule which estimates gradients with respect to the parameters of an underlying learning algorithm by simply running it twice. Our rule may be understood as a generalization of contrastive Hebbian learning to meta-learning and notably, it neither requires computing second derivatives nor going backwards in time, two characteristic features of previous gradient-based methods that are hard to conceive in physical neural circuits. We demonstrate the generality of our rule by applying it to two distinct models: a complex synapse with internal states which consolidate task-shared information, and a dual-system architecture in which a primary network is rapidly modulated by another one to learn the specifics of each task. For both models, our meta-learning rule matches or outperforms reference algorithms on a wide range of benchmark problems, while only using information presumed to be locally available at neurons and synapses. We corroborate these findings with a theoretical analysis of the gradient estimation error incurred by our rule.

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。

↑