论文标题
从示例中学习高级数学计算
Learning advanced mathematical computations from examples
论文作者
论文摘要
使用大型生成数据集上的变压器，我们训练模型来学习微分系统的数学属性，例如局部稳定性，无穷大的行为和可控性。我们几乎完美地预测了定性特征，以及系统的数值特征的良好近似值。这表明，神经网络可以学会在没有内置数学知识的情况下从示例中以高级理论为基础进行复杂的计算。
Using transformers over large generated datasets, we train models to learn mathematical properties of differential systems, such as local stability, behavior at infinity and controllability. We achieve near perfect prediction of qualitative characteristics, and good approximations of numerical features of the system. This demonstrates that neural networks can learn to perform complex computations, grounded in advanced theory, from examples, without built-in mathematical knowledge.
