What you need to know to train recurrent neural networks to make Flip Flops memories and more

Training neural networks to perform different tasks is relevant across various disciplines beyond Machine Learning. In particular, Recurrent Neural Networks (RNNs) are of great interest to different scientific communities. Open-source frameworks dedicated to Machine Learning, such as Tensorflow [1] and Keras [2] have produced significant changes in the development of technologies that we currently use. One relevant problem that can be approached with them is how to build the models to study dynamical systems and the brain. Specifically, how to extract the relevant information to answer the scientific questions of interest. The purpose of the present work is to contribute to this aim by analyzing a temporal processing task, in this case, a 3-bit Flip Flop memory. The modelling procedure in every step is shown: from equations to the software development. The networks obtained were analyzed to describe the dynamics and to show different visualization and analysis tools. The code developed in this premier is also provided to be used for modelling other tasks or systems.