The abstract submission deadline has been extended to take into account the decay of the decisions for submissions to the main conference.
The aim of this workshop is to bring together researchers and experts in machine learning, data mining, pattern analysis and statistics to share their challenges and advance research on temporal data analysis. Analysis and learning from temporal data cover a wide scope of tasks including learning metrics, learning representations, unsupervised feature extraction, clustering and classification.
This year, the workshop will be a hybrid event: organizers will welcome you in Grenoble, but it will also be possible to present your work, or attend the workshop remotely.
Temporal data are frequently encountered in a wide range of domains such as bioinformatics, medicine, finance and engineering, among many others. They are naturally present in applications covering language, motion and vision analysis, or more emerging ones as energy efficient building, smart cities, dynamic social media or sensor networks. Contrary to static data, temporal data are of complex nature, they are generally noisy, of high dimensionality, they may be non-stationary (i.e., first order statistics vary with time) and irregular (involving several time granularities), they may have several invariant domain-dependent factors as time delay, translation, scale or tendency effects. These temporal peculiarities are challenging for the majority of standard statistical models and machine learning approaches, that mainly assume i.i.d data, homoscedasticity, normality of residuals, etc. To tackle such challenging temporal data, we require new advanced approaches at the intersection of statistics, time series analysis, signal processing and machine learning. Defining new approaches that transcend boundaries between several domains to extract valuable information from temporal data is undeniably an important research topic, that has been the subject of active research in the last decade and will continue to do so for the foreseeable future.
AALTD welcomes papers that cover, but are not limited to, one or several of the following topics:
- Temporal data clustering
- Classification and regression of univariate and multivariate time series
- Early classification of temporal data
Deep learning for temporal data
- Learning representation for temporal data
- Metric and kernel learning for temporal data
- Modeling temporal dependencies
- Time series forecasting
- Time series annotation, segmentation and anomaly detection
- Spacial-temporal statistical analysis
- Functional data analysis methods
- Data streams
- Interpretable/explainable time-series analysis methods
- Dimensionality reduction, sparsity, algorithmic complexity and big data challenges
- Benchmarking and assessment methods for temporal data
- Applications, including bio-informatics, medical, energy consumption, on temporal data
We welcome contributions that address aspects including, but not limited to: novel techniques, innovative use and applications, techniques for the use of hybrid models. We are also considering inviting researchers and practitioners (including industrial) to present their time series management platforms and to raise open questions for which they do not find solutions off-the-shelf.