Adaptive Latent Space Tuning for Non-Stationary Distributions

Abstract

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.

References in corpus (21)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Deep learning for universal linear embeddings of nonlinear dynamics
• Identifying structural flow defects in disordered solids using machine learning methods
• Compression of sub-relativistic space-charge-dominated electron bunches for single-shot femtosecond electron diffraction
• Machine learning meets network science: dimensionality reduction for fast and efficient embedding of networks in the hyperbolic space
• Bayesian optimization of a free-electron laser
• Machine learning for beam dynamics studies at the CERN Large Hadron Collider
• Complex-valued neural networks for machine learning on non-stationary physical data
• Ultrafast Relativistic Electron Nanoprobes
• Accelerating lattice quantum Monte Carlo simulation using artificial neural networks: an application to the Holstein model
• Passive Ballistic Microbunching of Non-Ultrarelativistic Electron Bunches using Electromagnetic Wakefields in Dielectric-Lined Waveguides
• High-fidelity Prediction of Megapixel Longitudinal Phase-space Images of Electron Beams using Encoder-Decoder Neural Networks
• Exact representations of many body interactions with RBM neural networks
• Combining Machine Learning and Physics to Understand Glassy Systems
• Disentangling Trainability and Generalization in Deep Neural Networks
• Analysis of beam position monitor requirements with Bayesian Gaussian regression
• A general-purpose deep learning approach to model time-varying audio effects
• Extraction of instantaneous frequencies and amplitudes in nonstationary time-series data
• Reconstruction of Storage Ring 's Linear Optics with Bayesian Inference