Publications related to Data modeling

Match Normalization: Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World

In this work, we tackle the task of estimating the 6D pose of an object from point cloud data. While recent learning-based approaches have shown remarkable success on synthetic datasets, we have observed them to fail in the presence of real-world data. We ...

Ieee Computer Soc2024

High-Resolution Modeling Without Computation Slowdown for PETALE in CROCUS

Andreas Pautz, Vincent Pierre Lamirand, Thomas Jean-François Ligonnet, Axel Guy Marie Laureau

In a collaboration between Ecole Polytechnique Fédérale de Lausanne (EPFL) and CEA, in the fall of 2020, the experimental Programme d’Étude en Transmission de l’Acier Lourd et ses Eléments (PETALE) was successfully carried out in the CROCUS reactor of EPFL ...

2024

Generalization and Personalization of Machine Learning for Multimodal Mobile Sensing in Everyday Life

Lakmal Buddika Meegahapola

A range of behavioral and contextual factors, including eating and drinking behavior, mood, social context, and other daily activities, can significantly impact an individual's quality of life and overall well-being. Therefore, inferring everyday life aspe ...

EPFL2024

Extensions of Peer Prediction Incentive Mechanisms

Adam Julian Richardson

As large, data-driven artificial intelligence models become ubiquitous, guaranteeing high data quality is imperative for constructing models. Crowdsourcing, community sensing, and data filtering have long been the standard approaches to guaranteeing or imp ...

EPFL2024

Snowfall deposition in mountainous terrain: a statistical downscaling scheme from high-resolution model data on simulated topographies

Michael Lehning

One of the primary causes of non-uniform snowfall deposition on the ground in mountainous regions is the preferential deposition of snow, which results from the interaction of near-surface winds with topography and snow particles. However, producing high-r ...

Frontiers Media Sa2024

Quantifying the Unknown: Data-Driven Approaches and Applications in Energy Systems

Paul Scharnhorst

In light of the challenges posed by climate change and the goals of the Paris Agreement, electricity generation is shifting to a more renewable and decentralized pattern, while the operation of systems like buildings is increasingly electrified. This calls ...

EPFL2024

Models implemented in the methodological approach to design the initial STEP first wall contour

Davide Galassi, Kevin Henricus Annemarie Verhaegh, Filippo Bagnato, Dario Vaccaro

The official Spherical Tokamak for Energy Production mission aims to demonstrate the ability to generate net electricity from fusion with the STEP Prototype Power plant. One of the key technological and engineering challenges in fusion power plants is mana ...

Elsevier Science Sa2024

Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems

Colin Neil Jones, Roland Schwan, Melanie Nicole Zeilinger, Xuan Truong Nghiem

Physics-informed machine learning (PIML) is a set of methods and tools that systematically integrate machine learning (ML) algorithms with physical constraints and abstract mathematical models developed in scientific and engineering domains. As opposed to ...

IEEE2023

Blue Brain Nexus: An open, secure, scalable system for knowledge graph management and data-driven science

Henry Markram, Sean Lewis Hill, Samuel Claude Kerrien, Carolina Johanna Elisabeth Lindqvist, Alejandra Garcia Rojas Martinez, Huanxiang Lu, Mohameth François Sy, Anna-Kristin Kaufmann, Jonathan Raël Lurie, Henry Genet, Pierre-Alexandre Fonta, Alexander Désiré Ulbrich, Michaël Fernand Paul Dupont, Silvia Rosario Jimenez Tejeda, Bogdan Roman, Ian Lavriushev, Didac Montero Mendez, Wojciech Adam Wajerowicz, Kenneth William Pirman, Julien Antonin Machon, Dhanesh Neela Mana, Natalia Stafeeva

Modern data-driven science often consists of iterative cycles of data discovery, acquisition, preparation, analysis, model building and validation leading to knowledge discovery as well as dissemination at scale. The unique challenges of building and simul ...

2023

Stochastic Models for Comparison-based Search

Daniyar Chumbalov

In this thesis we study a problem of searching in a space of objects using comparisons. To navigate through the space to the target object

t

, we ask a sequence of questions of the form ``which object

i

or

j

is closer to

t

?'' for which we observe no ...

EPFL2023

Data-driven modeling of beam loss in the LHC

Tatiana Pieloni, Michael Schenk, Ekaterina Krymova, Loïc Thomas Davies Coyle

In the Large Hadron Collider, the beam losses are continuously measured for machine protection. By design, most of the particle losses occur in the collimation system, where the particles with high oscilla-tion amplitudes or large momentum error are scrape ...

FRONTIERS MEDIA SA2023

A comprehensive review of digital twin-part 2: roles of uncertainty quantification and optimization, a battery digital twin, and perspectives

Olga Fink, Chao Hu, Sayan Ghosh

As an emerging technology in the era of Industry 4.0, digital twin is gaining unprecedented attention because of its promise to further optimize process design, quality control, health monitoring, decision- and policy-making, and more, by comprehensively m ...

SPRINGER2023

E-Scan: Consuming Contextual Data with Model Plugins

Anastasia Ailamaki, Viktor Sanca

Extracting value and insights from increasingly heterogeneous data sources involves multiple systems combining and consuming the data. With multi-modal and context-rich data such as strings, text, videos, or images, the problem of standardizing the data mo ...

2023

Deriving Major Ion Concentrations at High Resolution from Continuous Electrical Conductivity Measurements in Karst Systems

Paolo Benettin

Time series of hydrochemical parameters support the investigation of dominant karst hydrological processes and conceptual model structures. Nevertheless, high costs for sample collection and analyses cause hydrochemical data to be rarely available at a suf ...

SPRINGER INTERNATIONAL PUBLISHING AG2023

Building a Knowledge Graph of Chinese Kung Fu Masters From Heterogeneous Bilingual Data

Lin Yuan, Yumeng Hou

Various endeavours into semantic web technologies and ontology engineering have been made within the organisation of cultural data, facilitating public access to digital assets. Although models for conceptualising objects have reached a certain level of ma ...

2023

Interactive-time Exploration, Querying, and Analysis of Large High-dimensional Datasets

Sachin Basil John

In the current era of big data, aggregation queries on high-dimensional datasets are frequently utilized to uncover hidden patterns, trends, and correlations critical for effective business decision-making. Data cubes facilitate such queries by employing p ...

EPFL2023

Modeling Structured Data in Attention-based Models

Alireza Mohammadshahi

Natural language processing has experienced significant improvements with the development of Transformer-based models, which employ self-attention mechanism and pre-training strategies. However, these models still present several obstacles. A notable issue ...

EPFL2023

Advances in Spillway Hydraulics: From Theory to Practice

Anton Schleiss, José Pedro Gamito de Saldanha Calado Matos

Over the past decades, significant advances have been achieved in hydraulic structures for dams, namely in water release structures such as spillway weirs, chutes, and energy dissipators. This editorial presents a brief overview of the eleven papers in thi ...

2023

TRACE: A Conceptual Model to Guide the Design of Educational Chatbots

Denis Gillet, Maria Jesus Rodriguez Triana, Juan Carlos Farah, Sandy Ingram, Fanny Kim-Lan Lasne, Adrian Christian Holzer

Driven by the rising popularity of chatbots such as ChatGPT, there is a budding line of research proposing guidelines for chatbot design, both in general and specifically for digital education. Nevertheless, few researchers have focused on providing concep ...

Springer2023

Harmonization of multi-scanner in vivo magnetic resonance spectroscopy: ENIGMA consortium task group considerations

Cristina Ramona Cudalbu

Magnetic resonance spectroscopy is a powerful, non-invasive, quantitative imaging technique that allows for the measurement of brain metabolites that has demonstrated utility in diagnosing and characterizing a broad range of neurological diseases. Its impa ...

FRONTIERS MEDIA SA2023