Self-supervised contrastive learning performs non-linear system identification
This work advances system identification for researchers in machine learning and control theory by providing a novel framework with theoretical backing.
The paper tackled the problem of system identification in latent space by showing that self-supervised contrastive learning can uncover linear, switching linear, and non-linear dynamics under a non-linear observation model, with theoretical guarantees and empirical validation.
Self-supervised learning (SSL) approaches have brought tremendous success across many tasks and domains. It has been argued that these successes can be attributed to a link between SSL and identifiable representation learning: Temporal structure and auxiliary variables ensure that latent representations are related to the true underlying generative factors of the data. Here, we deepen this connection and show that SSL can perform system identification in latent space. We propose dynamics contrastive learning, a framework to uncover linear, switching linear and non-linear dynamics under a non-linear observation model, give theoretical guarantees and validate them empirically.