Plasma state monitoring and disruption characterization using multimodal VAEs

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Recent progress in computer vision has been driven by a simple yet transformative observation: the potential of 30-year-old neural networks can be unlocked by massively scaling their parameters and the amount of human-labeled training data. This breakthrou ...

Semi-supervised semantic segmentation methods leverage unlabeled data by pseudo-labeling them. Thus the success of these methods hinges on the reliability of the pseudo-labels. Existing methods mostly choose high-confidence pixels in an effort to avoid err ...

Low-light photography produces images with low signal-to-noise ratios due to limited photons. In such conditions, common approximations like the Gaussian noise model fall short, and many denoising techniques fail to remove noise effectively. Although deep- ...

Motivation Many machine learning (ML) models developed to classify phenotype from gene expression data provide interpretations for their decisions, with the aim of understanding biological processes. For many models, including neural networks, interpretati ...

The evolution of airborne mapping witnesses the introduction of hybrid lidar-camera systems to enhance data collection, i.e. to obtain simultaneously high-density point-cloud and texture. Yet, the common adjustment of both optical data streams is a nontriv ...

Deep learning has shown remarkable potential for industrial applications, particularly in predictive maintenance and condition monitoring. A key challenge in this domain is prognostics, specifically predicting the Remaining Useful Life (RUL) of assets - a ...

Understanding the structure of real data is paramount in advancing modern deep-learning methodologies. Natural data such as images are believed to be composed of features organized in a hierarchical and combinatorial manner, which neural networks capture d ...

Power exhaust is a crucial issue for future fusion reactors. Divertor detachment and the required power dissipation fractions of about 95% are foreseen to be achieved by impurity seeding. In a tokamak, at high seeding levels the radiation often concentrate ...

Machine learning and artificial intelligence methods have achieved remarkable success, matching and even surpassing human capabilities in various complex tasks. However, many demonstrations have generally neglected a critical part of the intelligence that ...

Machine learning models (e.g., neural networks) achieve high accuracy in wind power forecasting, but they are usually regarded as black boxes that lack interpretability. To address this issue, the paper proposes a glass-box model that combines high accurac ...

Nuclear fusion provides a potentially unlimited and clean energy source which motivates the development of fusion power plants. The tokamak is one of the most promising concepts for a future reactor, confining the fusion plasmas in a torus using strong mag ...

Probing the neutral gas distribution at the edge of magnetic confinement fusion devices is critical for plasma exhaust studies. In the TCV tokamak, a set of ASDEX-type hot ionization pressure gauges (APGs) has been installed for fast, in situ measurements ...

Visual affordance segmentation identifies image regions of an object an agent can interact with. Existing methods re-use and adapt learning-based architectures for semantic segmentation to the affordance segmentation task and evaluate on small-size dataset ...

The density limit in strongly-shaped negative triangularity (NT) discharges is studied experimentally in the DIII-D tokamak. Record-high Greenwald fractions fG are obtained, using gas puff injection only, with values up to near 2, where fG is defined as th ...

This thesis focuses on understanding and improving the reasoning capabilities of neural networks. It develops theoretical results and empirical analyses to uncover reasoning potential and limitations, leveraging these insights to guide the design of improv ...

The problem of avoiding saturation of the coil currents is critical in large tokamaks with superconducting coils like ITER. Indeed, if the current limits are reached, a loss of control of the plasma may lead to a major disruption. Therefore, a current limi ...

When enough annotated training data is available, supervised deep-learning. algorithms excel at estimating human body pose and shape using a single camera. The effects of too little such data being available can be mitigated by using other information sour ...

In this thesis, we study two closely related directions: robustness and generalization in modern deep learning. Deep learning models based on empirical risk minimization are known to be often non-robust to small, worst-case perturbations known as adversari ...

Over the past years, image generation and manipulation have achieved remarkable progress due to the rapid development of generative AI based on deep learning. Recent studies have devoted significant efforts to address the problem of face image manipulation ...

The experimental fusion reactor ITER will feature two (or three) heating neutral beam injectors (NBI) capable of delivering 33(or 50) MW of power into the plasma. A NBI consists of a plasma source for production of negative ions (extracted negative ion cur ...

Plasma state monitoring and disruption characterization using multimodal VAEs

Chat with Graph Search

Exploiting Representation Similarities in Self-Supervised Learning for Vision Tasks

Weighting Pseudo-labels via High-Activation Feature Index Similarity and Object Detection for Semi-supervised Segmentation

Dark Noise Diffusion: Noise Synthesis for Low-Light Image Denoising

A comparative analysis of gene expression profiling by statistical and machine learning approaches

On the Perspectives of Image-to-Lidar Constraints in Dynamic Network Optimisation

Overcoming Distribution Shifts and Imbalance Challenges in Representation Learning for Deep Regression Models

A phase transition in diffusion models reveals the hierarchical nature of data

X-point radiation: From discovery to potential application in a future reactor

Sequential Reasoning with Socially Caused Beliefs

Explainable modeling for wind power forecasting: A Glass-Box model with high accuracy

Impurity-seeded detachment in conventional and optimized TCV divertor geometry

Neutral pressure measurement in TCV tokamak using ASDEX-type pressure gauges

Segmenting Object Affordances: Reproducibility and Sensitivity to Scale

Operation Above the Greenwald Density Limit in High Performance DIII-d Negative Triangularity Discharges

Learning to Reason with Neural Networks: Principles and Methods

Implementation of an Iter-relevant Qp-based Current Limit Avoidance Algorithm in the Tcv Tokamak

Using Motion Cues to Supervise Single-Frame Body Pose and Shape Estimation in Low Data Regimes

Understanding generalization and robustness in modern deep learning

Towards the Detection of AI-Synthesized Human Face Images

Optimizing the ITER NBI ion source by dedicated RF driver test stand