tension: A Python package for FORCE learning.

tension: A Python package for FORCE learning.
复制标题

DOI:
10.1371/journal.pcbi.1010722
复制
发表时间:
2022-12
影响因子:
4.3
通讯作者:
--
中科院分区:
生物学2区
文献类型:
--
作者:

文献摘要

参考文献

相似文献

一阶,减少和受控的误差(力)学习及其变体被广泛用于训练混乱的复发性神经网络(RNN),并且在某些任务上的梯度方法优于梯度方法。但是,目前尚无针对力学习的标准软件框架。我们提出了张力,这是一种面向对象的开源python软件包,它实现了张量 / keras api的力。我们展示了如何使用共享的,易于扩展的高级API对受生物数据限制的速率网络,尖峰网络和网络都可以培训。借助相同的资源,我们的实施在运行时的损失和发表的力量实施方面优于常规RNN。我们在这里的工作使力量训练混乱的RNN可访问和简单地迭代,并促进了对感兴趣的行为如何从神经动力学中产生的建模。
First-Order, Reduced and Controlled Error (FORCE) learning and its variants are widely used to train chaotic recurrent neural networks (RNNs), and outperform gradient methods on certain tasks. However, there is currently no standard software framework for FORCE learning. We present tension, an object-oriented, open-source Python package that implements a TensorFlow / Keras API for FORCE. We show how rate networks, spiking networks, and networks constrained by biological data can all be trained using a shared, easily extensible high-level API. With the same resources, our implementation outperforms a conventional RNN in loss and published FORCE implementations in runtime. Our work here makes FORCE training chaotic RNNs accessible and simple to iterate, and facilitates modeling of how behaviors of interest emerge from neural dynamics.
运动规划期间,前皮层的强大神经元动力学。
DOI: 10.1038/nature17643
发表时间: 2016-04-28
期刊: NATURE
影响因子: 64.8
作者:
Li, Nuo;Daie, Kayvon;Svoboda, Karel;Druckmann, Shaul
通讯作者: Druckmann, Shaul
DOI: 10.1038/s41467-017-01827-3
发表时间: 2017-12-20
影响因子: 16.6
作者:
Nicola W;Clopath C
通讯作者: Clopath C
DOI: 10.1016/j.cosrev.2009.03.005
发表时间: 2009-08-01
影响因子: 12.9
作者:
Lukosevicius, Mantas;Jaeger, Herbert
通讯作者: Jaeger, Herbert
DOI: 10.1016/j.neuron.2009.07.018
发表时间: 2009-08-27
期刊: NEURON
影响因子: 16.2
作者:
Sussillo, David;Abbott, L. F.
通讯作者: Abbott, L. F.
DOI: 10.1038/s41598-018-24271-9
发表时间: 2018-04-17
期刊: Scientific reports
影响因子: 4.6
作者:
Che Z;Purushotham S;Cho K;Sontag D;Liu Y
通讯作者: Liu Y