Related papers: Outlier classification using Autoencoders: applica…
An analysis of plasma density distributions at arbitrary ion-atom collisionality for one-dimensional axially symmetrical cylindrical and annular plasmas is presented. Perturbations of plasma densities caused by a cylindrical probe are…
Data-driven reduced-order models based on autoencoders generally lack interpretability compared to classical methods such as the proper orthogonal decomposition. More interpretability can be gained by disentangling the latent variables and…
This paper examines the problem of locating outlier columns in a large, otherwise low-rank, matrix. We propose a simple two-step adaptive sensing and inference approach and establish theoretical guarantees for its performance; our results…
Multi-label (ML) classification is an actively researched topic currently, which deals with convoluted and overlapping boundaries that arise due to several labels being active for a particular data instance. We propose a classifier capable…
Deep networks often make confident, yet, incorrect, predictions when tested with outlier data that is far removed from their training distributions. Likelihoods computed by deep generative models (DGMs) are a candidate metric for outlier…
We report on a recently developed laser-probing diagnostic which allows direct measurements of ray-deflection angles in one axis, whilst retaining imaging capabilities in the other axis. This allows us to measure the spectrum of angular…
The asymmetric simple exclusion process with additional Langmuir kinetics, i.e. attachment and detachment in the bulk, is a paradigmatic model for intracellular transport. Here we study this model in the presence of randomly distributed…
We derive the statistics of scattered photons from a $\Lambda$- or ladder-type three-level emitter (3LE) embedded in a 1D open waveguide. The weak probe photons in the waveguide are coupled to one of the two allowed transitions of the 3LE,…
Latent diffusion models (LDMs) power state-of-the-art high-resolution generative image models. LDMs learn the data distribution in the latent space of an autoencoder (AE) and produce images by mapping the generated latents into RGB image…
We present an exact calculation of the power spectrum of the electromagnetic fluctuations in a relativistic equilibrium plasma described by Maxwell-J\"uttner distribution functions. We consider the cases of wave vectors parallel or normal…
We develop Riemannian approaches to variational autoencoders (VAEs) for PDE-type ambient data with regularizing geometric latent dynamics, which we refer to as VAE-DLM, or VAEs with dynamical latent manifolds. We redevelop the VAE framework…
An extended thermal particle boundary condition is devised to more efficiently and accurately model laser-plasma interactions in overdense plasmas. Particle-in-cell simulations of such interactions require many particles per cell, and a…
It is important to detect anomalous inputs when deploying machine learning systems. The use of larger and more complex inputs in deep learning magnifies the difficulty of distinguishing between anomalous and in-distribution examples. At the…
Electron and scanning probe microscopy produce vast amounts of data in the form of images or hyperspectral data, such as EELS or 4D STEM, that contain information on a wide range of structural, physical, and chemical properties of…
Outlying observations are frequently encountered across a wide spectrum of scientific domains, posing notable challenges to the generalizability of statistical models and the reproducibility of downstream analysis. They are identified…
Human operators often diagnose industrial machinery via anomalous sounds. Automated acoustic anomaly detection can lead to reliable maintenance of machinery. However, deep learning-driven anomaly detection methods often require an extensive…
Many of the existing machine learning algorithms, both supervised and unsupervised, depend on the quality of the input characteristics to generate a good model. The amount of these variables is also important, since performance tends to…
Deep learning has made significant strides in medical imaging, leveraging the use of large datasets to improve diagnostics and prognostics. However, large datasets often come with inherent errors through subject selection and acquisition.…
The self-consistent description of Langmuir wave and ion-sound wave turbulence in the presence of an electron beam is presented for inhomogeneous non-isothermal plasmas. Full numerical solutions of the complete set of kinetic equations for…
Thanks to advances in plasma science and enabling technology, mirror machines are being reconsidered for fusion power plants and as possible fusion volumetric neutron sources. However cross-field transport and turbulence in mirrors remains…