Mathematical Foundations of Control Driven by Large Data

in Special Issue   Posted on April 29, 2021 

Information for the Special Issue

Submission Deadline: Sun 31 Oct 2021
Journal Impact Factor : 1.667
Journal Name : Mathematics of Control, Signals, and Systems
Journal Publisher:
Website for the Special Issue: https://www.springer.com/journal/498/updates/19042540?error=cookies_not_supported&code=1a5575bd-57a0-4a60-aace-80fe486d7e92
Journal & Submission Website: https://www.springer.com/journal/498

Special Issue Call for Papers:

A Special Issue in Mathematics of Control, Signals, and SystemsOpen for submissions until October 31 2021

Recent advances in sensing and data collection capabilities provide access to exponentially increasing amounts of data. This data availability opens up a wide range of applications — from safer, self-aware environments, and smart cities to autonomous vehicles — that have the potential to profoundly impact society. However, in order to realize this potential, new control algorithms are needed, capable of handling large amounts of data, often in real time, using computational and power budgets compatible with on-board resources. While this is beyond the capability of “traditional” control algorithms, during the past few years, new control methods have started to appear specifically tailored to handle big data, for instance by exploiting the structure of the problem or by adapting methods from other communities such as machine learning, sparse optimization, and randomized linear algebra. Motivated by these developments, MCSS is inviting mathematically rigorous original research articles for a special issue on the Mathematical Foundations of Control Driven by Large Data.

This special issue welcomes papers on topics that include but are not limited to:

  • Learning controllers directly from data
  • Systems identification from large data sets
  • Sample complexity of learning dynamical systems
  • Data driven optimization of dynamical systems
  • Machine learning inspired approaches to learning and controlling dynamical systems, including but not limited to Reinforcement Learning.
  • Use of control theoretic techniques to improve learning, including convergence rates and robustness against adversarial attacks.
  • Handling different scales of data availability and informativity.

Submission detailsSubmissions should be made beginning August 1st 2021 until October 31, 2021, and will undergo a rigorous peer-review process. The issue is published in the form of a topical collection that will be available online as soon as the first paper is accepted. For further author guidelines, please refer to https://www.springer.com/journal/498/submission-guidelines.

For any questions, please contact one of the guest editors:

Alessandro Chiuso [email protected] Ozay [email protected] Sznaier [email protected] Vidal [email protected]

Other Special Issues on this journal

Closed Special Issues