Statistical Signal Processing Solutions and Advances for Data Science: Complex, Dynamic and Large-scale Settings
May 1, 2019 (closed)
Statistical Signal Processing has faced new challenges and a paradigm shift towards data science due to technological increase in computational power, explosion in number of connected devices in the internet and the ever increasing amounts of data volumes generated by today’s ubiquitous communication, imaging, e-commerce and social media. Consequently new approaches, methods, theory and tools are developed by signal processing community to account for modern complex, dynamic and large scale settings with complex yet hidden low-dimensional underlying structures.
This special issue will provide a modern look on recent trends and advances on statistical signal processing towards data science that account for a) complexity of the data which can be represented as low rank structures and subspaces, sparsity and missing values, or due to sheer variety of the data b) large scale settings which refers to high-dimensionality but also to the settings where sample size is smaller or not much larger than the dimension and hence make asymptotically optimal methods perform poorly and c) dynamic nature of the data which accumulates or streams at fast pace.
Topics of interest include:
- Prospective authors are invited to submit high-quality original contributions and reviews for this Special Issue. Potential topics include, but are not limited to:
- * random matrix theory
- * large-scale statistical inference and learning
- * robust statistics
- * large-scale optimization and optimization on manifolds
- * regularization techniques and sparsity-driven approaches
- * new representations and models to handle such data structures including graph signal processing, tensor data analysis and multi-linear algebra, latent-variable analysis models, and sparse signal representations and dictionaries.
- Esa Ollila, Department of Signal Processing and Acoustics, Aalto University, Finland
- Michael Muma, Department of Signal Processing, Technische Universität Darmstadt, Germany
- Frédéric Pascal, L2S, Centrale Supélec, France