Learning latent structure: carving nature at its joints.

Learning latent structure: carving nature at its joints.
复制标题

DOI:
10.1016/j.conb.2010.02.008
复制
发表时间:
2010-04
影响因子:
5.7
通讯作者:
Niv Y
Niv Y
中科院分区:
医学2区
文献类型:
--
作者:
Gershman SJ;Niv Y

文献摘要

参考文献

被引文献

相似文献

强化学习算法为简单的学习和决策行为及其底层神经基质的功能提供了强有力的解释。不幸的是,在涉及许多刺激和行动的现实世界中,这些算法的学习速度慢得可怜,与动物和人类的学习相比,它们的学习能力很差。在这里,我们认为这种差异的一个原因是人类和动物利用现实世界任务固有的结构来简化学习问题。我们调查了关于“结构学习”的新兴文献-使用经验来推断任务的结构-以及这如何有助于强化学习,重点是感知和行动中的结构。
Reinforcement learning algorithms provide powerful explanations for simple learning and decision making behaviors and the functions of their underlying neural substrates. Unfortunately, in real world situations that involve many stimuli and actions, these algorithms learn pitifully slowly, exposing their inferiority in comparison to animal and human learning. Here we suggest that one reason for this discrepancy is that humans and animals take advantage of structure that is inherent in real-world tasks to simplify the learning problem. We survey an emerging literature on “structure learning”—using experience to infer the structure of a task—and how this can be of service to reinforcement learning, with an emphasis on structure in perception and action.
DOI: 10.1037/a0017808
发表时间: 2010-01-01
影响因子: 5.4
作者:
Gershman, Samuel J.;Blei, David M.;Niv, Yael
通讯作者: Niv, Yael
DOI: 10.1016/j.biopsych.2008.05.009
发表时间: 2008-11-01
影响因子: 10.6
作者:
Ceaser, Alan E.;Goldberg, Terry E.;Gold, James M.
通讯作者: Gold, James M.
在行动中构建学习。
DOI: 10.1016/j.bbr.2009.08.031
发表时间: 2010-01-20
影响因子: 2.7
作者:
Braun, Daniel A.;Mehring, Carsten;Wolpert, Daniel M.
通讯作者: Wolpert, Daniel M.
DOI: 10.1016/s0006-3223(99)00116-x
发表时间: 1999-08-01
影响因子: 10.6
作者:
Braver, TS;Barch, DM;Cohen, JD
通讯作者: Cohen, JD
DOI: 10.1037/a0014282
发表时间: 2009-01-01
影响因子: 5.4
作者:
Kemp, Charles;Tenenbaum, Joshua B.
通讯作者: Tenenbaum, Joshua B.