Provably Correct Intelligent Learning CPS
Claire Tomlin

Citation
Claire Tomlin. "Provably Correct Intelligent Learning CPS". Talk or presentation, 23, August, 2017.

Abstract
The goal of this project is to develop design tools for CPS systems which have elements of machine learning incorporated into them. By this we mean the provably correct design of learning systems that are making their appearance in autonomous and semi-autonomous systems (self-driving cars, drones, etc.). The project is ambitious in bringing together progress in the design of provably correct embedded CPS systems with machine learning and data analytics from the Internet of Things. By provably correct we mean probabilistic guarantees on the performance of systems with machine learning approaches built in. The potential benefits of the project to the development of model based design tools for intelligent systems are huge. Learning is now ubiquitously intertwined in self-driving cars, drones, wireless platforms, brain machine interfaces such as hearing aids, and medical devices. The performance of closed loop learning systems is however only modestly predictable. By Provably Correct, we mean that we are able to provide probabilistic rather than deterministic guarantees for the performance of the closed loop system, leading to what we term HiCLAS-CPS, for High Confidence Learning and Adaptive Systems for CPS.

Electronic downloads


(No downloads are available for this publication.)
Citation formats  
  • HTML
    Claire Tomlin. <a
    href="http://www.cps-forces.org/pubs/260.html"
    ><i>Provably Correct Intelligent Learning
    CPS</i></a>, Talk or presentation,  23, August,
    2017.
  • Plain text
    Claire Tomlin. "Provably Correct Intelligent Learning
    CPS". Talk or presentation,  23, August, 2017.
  • BibTeX
    @presentation{Tomlin17_ProvablyCorrectIntelligentLearningCPS,
        author = {Claire Tomlin},
        title = {Provably Correct Intelligent Learning CPS},
        day = {23},
        month = {August},
        year = {2017},
        abstract = {The goal of this project is to develop design
                  tools for CPS systems which have elements of
                  machine learning incorporated into them. By this
                  we mean the provably correct design of learning
                  systems that are making their appearance in
                  autonomous and semi-autonomous systems
                  (self-driving cars, drones, etc.). The project is
                  ambitious in bringing together progress in the
                  design of provably correct embedded CPS systems
                  with machine learning and data analytics from the
                  Internet of Things. By provably correct we mean
                  probabilistic guarantees on the performance of
                  systems with machine learning approaches built in.
                  The potential benefits of the project to the
                  development of model based design tools for
                  intelligent systems are huge. Learning is now
                  ubiquitously intertwined in self-driving cars,
                  drones, wireless platforms, brain machine
                  interfaces such as hearing aids, and medical
                  devices. The performance of closed loop learning
                  systems is however only modestly predictable. By
                  Provably Correct, we mean that we are able to
                  provide probabilistic rather than deterministic
                  guarantees for the performance of the closed loop
                  system, leading to what we term HiCLAS-CPS, for
                  High Confidence Learning and Adaptive Systems for
                  CPS.},
        URL = {http://cps-forces.org/pubs/260.html}
    }
    

Posted by Carolyn Winter on 24 Aug 2017.
Groups: forces
For additional information, see the Publications FAQ or contact webmaster at cps-forces org.

Notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright.