The workshop will be held on Monday, September 13, 2021. It will be organized fully virtually within the ECML conference system and using Zoom.
The main sessions of the workshop will be organized using the Zoom platform. Please take care of the Zoom version: we recommend to update it to be sure to be able to navigate through the different rooms during the poster session.
All the connection information will be available on the ECML Whova program manager (please, note that attending this workshop is not free of charge. A registration to ECML conference is required). Access to the AALTD Whova page, go there.
All papers can be downloaded on the accepted papers page.
(local time at Bilbao, Spain: CEST = UTC+2)
- 9h30 – 9h45: Workshop introduction by organizers Live session on Zoom
- 9h45 – 10h45: Session Time Series: Classification and Forecasting
- Alessio Benavoli and Giorgio Corani. State Space approximation of Gaussian Processes for time-series forecasting
- Bhaskar Dhariyal, Thach Le Nguyen and Georgiana Ifrim. Fast Channel Selection for Scalable Multivariate Time Series Classification
- Surabhi Agarwal, Trang Thu Nguyen, Thach Le Nguyen and Georgiana Ifrim. Ranking by Aggregating Referees: Evaluating the Informativeness of Explanation Methods for Time Series Classification
- Break
- 11h – 12h20: Session Time Series: Applications
- Mathieu Chambard, Thomas Guyet, Yên-Lan Nguyen and Etienne Audureau Temporal phenotyping for characterization of hospital care pathways of COVID patients
- Etienne Goffinet, Mustapha Lebbah, Hanane Azzag, Loïc Giraldi and Anthony Coutant. A New Multivariate Time Series Co-clustering Non-Parametric Model Applied to Driving-Assistance Systems Validation
- Cristian Axenie, Daniele Foroni, Alexander Wieder, Mohamad Al Hajj Hassan, Paolo Sottovia, Margherita Grossi, Rongye Shi, Stefano Bortoli and Götz Brasche. TRAMESINO: Trainable Memory System for Intelligent Optimization of Road Traffic Control
- Laurens Stoop, Erik Duijm, Ad Feelders and Machteld van den Broek. Detection of critical events in renewable energy production time series
- Lunch break
- 14h30 – 15h : Spotlight presentations of the posters (5 minutes per poster, the list of posters is here)
- 15h – 16h: Virtual poster session.
- Sebastian Pineda Arango, Felix Heinrich, Kiran Madhusudhanan and Lars Schmidt-Thieme. Multimodal Meta-Learning for Time Series Regression
- Tom van de Looij and Mozhdeh Ariannezhad. Cluster-based Forecasting for Intermittent and Non-intermittent Time Series
- Olli-Pekka Rinta-Koski, Miki Sirola, Le Ngu Nguyen and Jaakko Hollmén. State discovery and prediction from multivariate sensor data
- Abdul Hameed Azeemi, Muhammad Hamza Sohail, Talha Zubair, Muaz Maqbool, Irfan Younas and Omair Shafiq. RevDet: Robust and Memory Efficient Event Detection and Tracking in Large News Feeds
- Julien Audibert, Sébastien Marti, Frédéric Guyard and Maria A. Zuluaga. From Univariate to Multivariate Time Series Anomaly Detection with Non-Local Information
- Break
- 17h30 – 18h15: Invited talk Dr Rose Yu Deep Generative Models for Missing Data in Temporal Sequences (abstract)
- 18h30 Workshop Closing
