MLVis 2021 logo
 

MLVis 2021, International Workshop on Machine Learning in Visualisation for Big Data 2021
Co-located with Eurovis 2021, June 14, 2021, Online

Update: we have extended the submission deadline to April 16, and the notification deadline to May 7.

Machine Learning Methods in Visualisation will be held as part of Eurovis 2021. The sixth edition of this co-located event will be part-tutorial and part-workshop so as to increase the interaction between researchers.

Part of the MLVis 2021 programme will consist of short papers.

We solicit short papers on machine learning methods in visualisation from both the machine learning and visualisation communities, addressing how the two technologies can be used together to provide greater insight to end users.

Topics include but are not limited to: Explainable Machine Learning, Dimensionality Reduction, Visualisation of Clustering, Regression, and Classification, Steerable Machine Learning, Visualisation to Improve Machine Learning Models, Automation of Visualisation and Visual Analytics, Visualisation and Machine Learning in Text Analytics, Visualisation in Online Machine Learning

Important Dates

  • Submission deadline (extended): April 16, 2021
  • Notification deadline (extended): May 7, 2021
  • Camera-ready deadline: May 14, 2021

All submission deadlines are at 23:59 Anywhere on Earth on the date indicated.

Submission Guidelines

Paper submissions for MLVis should be at most 4 pages in the MLVis 2021 latex style, with an additional page allowed for references. All submissions must be original works that have not been published previously in any conference proceedings, magazine, journal, or edited book. Papers are to be submitted via the new PCS at: https://new.precisionconference.com (choose society “Eurographics” and conference/journal “MLVis 2021” when making your submission). The submissions do not need to be blind.

Publication

At least one author of an accepted paper must register and participate in the MLVis 2021 workshop to present the accepted work. MLVis will be organized online, co-located with Eurovis 2021.

Proceedings will be published by the Eurographics Association, and be stored on the Eurographics Digital Library.

Selected papers may be invited to ACM TiiS.

Arrangements

The workshop will be broadcast on YouTube Live, and comments can be made either in the YouTube Live chat or in a Discord channel.

We encourage particants to use Discord for more lengthy discussions. The link to the MLVis 2021 discord channel will be provided soon.

Programme

Part 1 (14:00-15:35 CEST, chair: Daniel Archambault)
Links: YouTube Live stream, Discord chat channel
  • 14:00 - 14:05 CEST: Opening, live component
  • 14:05 - 15:05 CEST: Keynote - Torsten Möller, Universität Wien
    Visually supported exploration of multiple objectives
    When we build models, in Data Science or in Computational Science, we need to specify an objective function to find (or learn) a best model. However, in many scenarios there are multiple objectives one has to consider. Often times, we simply specify a weighted average among relevant objective functions. In this talk I will report on how to deal with cases where such a weighting is not clear or possible: how can we deal with multiple objectives in model building? By creating a visual analysis workflow that complements algorithmic and mathematical analysis workflows, I will try to convince you that we are able to better understand multi-dimensional problems and solution spaces.
  • 15:05 - 15:15 CEST: Ian Nabney, live tutorial component
    Visualisation of Mixed Data Types
  • 15:15 - 15:25 CEST: Jaakko Peltonen, live tutorial component
    Dimensionality Reduction and Visual Interfaces
  • 15:25 - 15:35 CEST: Daniel Archambault, live tutorial component
    Visual Analytics Solutions to Compartmental Modelling in Response to COVID-19

15:45 - 16:00 CEST: Break

Part 2 (16:00 - 17:40 CEST): Papers and Panel (chair: Ian Nabney)
Links: YouTube Live stream, Discord chat channel
  • 16:00 - 16:25 CEST: Controllably Sparse Perturbations of Robust Classifiers for Explaining Predictions and Probing Learned Concepts
    Jay Roberts, Theodoros Tsiligkaridis
  • 16:25 - 16:50 CEST: Revealing Multimodality in Ensemble Weather Prediction
    Natacha Galmiche, Helwig Hauser, Thomas Spengler, Clemens Spensberger, Morten Brun, Nello Blaser
  • 16:50 - 17:30 CEST: Live Panel with Presenters
  • 17:30 - 17:40 CEST: Closing

Organising Committee

picture of Daniel Archambault Daniel Archambault
(Swansea University, United Kingdom)
picture of Ian Nabney Ian Nabney
(University of Bristol, United Kingdom)
picture of Jaakko Peltonen Jaakko Peltonen
(Tampere University, Finland)

International Program Committee

Previous Events