Codalab competitions: An open source platform to organize scientific challenges

A Pavao, I Guyon, AC Letournel, DT Tran… - Journal of Machine …, 2023 - jmlr.org
CodaLab Competitions is an open source web platform designed to help data scientists and
research teams to crowd-source the resolution of machine learning problems through the …

The ENIGMA Stroke Recovery Working Group: Big data neuroimaging to study brain–behavior relationships after stroke

SL Liew, A Zavaliangos‐Petropulu… - Human brain …, 2022 - Wiley Online Library
The goal of the Enhancing Neuroimaging Genetics through Meta‐Analysis (ENIGMA) Stroke
Recovery working group is to understand brain and behavior relationships using well …

Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?

B Kégl, G Hurtado, A Thomas - arXiv preprint arXiv:2107.11587, 2021 - arxiv.org
We contribute to micro-data model-based reinforcement learning (MBRL) by rigorously
comparing popular generative models using a fixed (random shooting) control agent. We …

Evolution of the historian data entry application: Supporting transcribathons in the digital humanities through mdd

A Schieweck, R Murphy, R Khan… - 2022 IEEE 46th …, 2022 - ieeexplore.ieee.org
Death and Burial Data: Ireland 1864–1922 (DBDIrl), is a digital humanities project, which
uses historical civil registration of death as its primary dataset. The overarching aim of this …

A Multi-step Loss Function for Robust Learning of the Dynamics in Model-based Reinforcement Learning

A Benechehab, A Thomas, G Paolo… - arXiv preprint arXiv …, 2024 - arxiv.org
In model-based reinforcement learning, most algorithms rely on simulating trajectories from
one-step models of the dynamics learned on data. A critical challenge of this approach is the …

Sharing and performance optimization of reproducible workflows in the cloud

R Qasha, Z Wen, J Cała, P Watson - Future Generation Computer Systems, 2019 - Elsevier
Scientific workflows play a vital role in modern science as they enable scientists to specify,
share and reuse computational experiments. To maximizethe benefits, workflows need to …

The 2018 climate informatics hackathon: Hurricane intensity forecast

S Giffard-Roisin, D Gagne, A Boucaud, B Kégl… - … Workshop on Climate …, 2018 - hal.science
The 2018 Climate Informatics hackathon focused on forecasting the hurricane intensities,
and there was 35 participants. Specifically, the goal was to predict the intensity of tropical …

AI Competitions and Benchmarks: How to ensure a long-lasting impact of a challenge with post-challenge paper, benchmarks and other dissemination action

A Marot, D Rousseau, Z Xu - arXiv preprint arXiv:2312.06036, 2023 - arxiv.org
Organising an AI challenge does not end with the final event. The long-lasting impact also
needs to be organised. This chapter covers the various activities after the challenge is …

Reproducing scientific experiment with cloud devops

F Zhao, X Niu, SL Huang… - 2020 IEEE World Congress …, 2020 - ieeexplore.ieee.org
The reproducibility of scientific experiment is vital for the advancement of disciplines based
on previous work. To achieve this goal, many researchers focus on complex methodology …

Recent Advances in Tropical Cyclone Forecasting Using Machine Learning on Reanalysis and Remote Sensing

S GIFFARD‐ROISIN - Multitemporal Earth Observation Image …, 2024 - books.google.com
In this chapter, we will show how the recent advances of machine learning (ML) and in
particular deep learning (DL) can be applied to one major natural hazard forecasting …