Adaptive Latent Space Tuning for Non-Stationary Distributions
This addresses the problem of non-stationary distributions in deep learning for real-time applications like particle accelerators, but it is incremental as it builds on existing encoder-decoder CNN frameworks.
The paper tackles the challenge of predicting properties in time-varying systems with fast distribution shifts, such as a particle accelerator's charged particle beam, by introducing an adaptive tuning method for the latent space of encoder-decoder CNNs, achieving real-time compensation without retraining.
Powerful deep learning tools, such as convolutional neural networks (CNN), are able to learn the input-output relationships of large complicated systems directly from data. Encoder-decoder deep CNNs are able to extract features directly from images, mix them with scalar inputs within a general low-dimensional latent space, and then generate new complex 2D outputs which represent complex physical phenomenon. One important challenge faced by deep learning methods is large non-stationary systems whose characteristics change quickly with time for which re-training is not feasible. In this paper we present a method for adaptive tuning of the low-dimensional latent space of deep encoder-decoder style CNNs based on real-time feedback to quickly compensate for unknown and fast distribution shifts. We demonstrate our approach for predicting the properties of a time-varying charged particle beam in a particle accelerator whose components (accelerating electric fields and focusing magnetic fields) are also quickly changing with time.