Turbo-Sim: a generalised generative model with a physical latent space
This work provides a general framework for generative modeling, potentially benefiting researchers in machine learning and physics, but it appears incremental as it builds upon and unifies existing methods without claiming broad SOTA improvements.
The authors tackled the problem of creating a mathematically interpretable generative model by introducing Turbo-Sim, a generalized autoencoder framework derived from information theory, which unifies and extends existing models like adversarial autoencoders and GANs, and demonstrated its application in collider physics for particle property transformation.
We present Turbo-Sim, a generalised autoencoder framework derived from principles of information theory that can be used as a generative model. By maximising the mutual information between the input and the output of both the encoder and the decoder, we are able to rediscover the loss terms usually found in adversarial autoencoders and generative adversarial networks, as well as various more sophisticated related models. Our generalised framework makes these models mathematically interpretable and allows for a diversity of new ones by setting the weight of each loss term separately. The framework is also independent of the intrinsic architecture of the encoder and the decoder thus leaving a wide choice for the building blocks of the whole network. We apply Turbo-Sim to a collider physics generation problem: the transformation of the properties of several particles from a theory space, right after the collision, to an observation space, right after the detection in an experiment.