Adaptive Latent Space Tuning for Non-Stationary Distributions
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 17 of its 89 references that are in the map. First posted on arXiv 8 May 2021.
Builds on
17 of 89 references.
References · 17
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2015Spatial Transformer Networkscited by 1,423
- 2021Going deeper with Image Transformerscited by 759
- 2016TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systemscited by 660
- 2017Deep Neural Networks as Gaussian Processescited by 617
- 2017Deep learning for universal linear embeddings of nonlinear dynamicscited by 491
- 2020AdapterHub: A Framework for Adapting Transformerscited by 441
- 2019Machine Learning for Fluid Mechanicscited by 380
- 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 287
- 2016Theano: A Python framework for fast computation of mathematical expressionscited by 64
- 2016Machine learning meets network science: dimensionality reduction for fast and efficient embedding of networks in the hyperbolic spacecited by 21
- 2019A general-purpose deep learning approach to model time-varying audio effectscited by 8
5 more
- 2020Adaptive 3D convolutional neural network-based reconstruction method for 3D coherent diffraction imagingcited by 8
- 2019Complex-valued neural networks for machine learning on non-stationary physical datacited by 7
- 2020Accurate and confident prediction of electron beam longitudinal properties using spectral virtual diagnosticscited by 5
- 2021Extraction of instantaneous frequencies and amplitudes in nonstationary time-series datacited by 3
- 2017Combining Machine Learning and Physics to Understand Glassy Systemscited by 1
Built on by
Citing papers per year: 2021: 1.
1 paper.