1: \begin{abstract}

2: 		\hs\\

3: 		Ordinary Differential Equations are generally too complex to be solved analytically. Approximations thereof  can be obtained by general purpose numerical methods. However, even though accurate schemes have been developed, they remain computationally expensive: In this paper, we resort to the theory of modified equations in order to obtain "on the fly"  cheap numerical approximations. The recipe consists in approximating, prior to that, the modified field associated to the modified equation by {\em neural networks}.  Elementary convergence results  are then  established and the efficiency of the technique is demonstrated on experiments.

4: 	\end{abstract}

5: