Random DFAs and RNNs

该数据集包含30个确定有限自动机（DFAs）和循环神经网络（RNNs），同时提供了138个规范用于评估模型检查算法。这些循环神经网络是在一个包含5个字母的字母表中的单词上进行训练的，训练过程确保了RNN在训练集上的准确率超过95%。数据集的规模包括30个DFAs/RNNs和138个规范，其任务是使用模型检查算法来验证循环神经网络。

This dataset contains 30 Deterministic Finite Automata (DFAs) and Recurrent Neural Networks (RNNs), along with 138 specifications for evaluating model checking algorithms. These recurrent neural networks were trained on words from a 5-letter alphabet, and the training procedure ensured that the RNNs achieved an accuracy exceeding 95% on the training set. With 30 DFAs/RNNs and 138 specifications in total, the core task of this dataset is to verify recurrent neural networks using model checking algorithms.