We studied the properties of simple recurrent neural networks trained to perform temporal tasks and also flow control tasks with temporal stimulus. We studied mainly three aspects: inner configuration sets, memory capacity with the scale of the models and finally immunity to induced damage on a trained network. Our results allow us to quantify different aspects of these models which are normally used as black boxes to model the biological response of cerebral cortex.

翻译：我们研究了简单的经常性神经网络的特性,这些网络受过训练,可以执行时间性任务,还可以通过时间性刺激进行流动控制任务。我们主要研究了三个方面:内构件、模型规模的记忆能力以及最终在经过培训的网络上免受诱发损害。我们的结果使我们能够量化这些模型的不同方面,这些模型通常被用作黑盒,用来模拟大脑皮层的生物反应。