1: \begin{abstract}

2:   Likelihood-free inference is quickly emerging as a powerful tool to perform fast/effective parameter estimation.

3:   We demonstrate a technique of optimizing likelihood-free inference to make it even faster by marginalizing symmetries

4:   in a physical problem. In this approach, physical symmetries, for example, time-translation are learned using joint-embedding

5:   via self-supervised learning with symmetry data augmentations. Subsequently, parameter inference is performed using a normalizing flow where

6:   the embedding network is used to summarize the data before conditioning the parameters. We present this approach on two simple

7:   physical problems and we show faster convergence in a smaller number of parameters compared to a normalizing flow that does not use a pre-trained symmetry-informed

8:   representation.

9: \end{abstract}

10: