Call for Papers/Participation

NIPS*2001 Workshop on

Artificial Neural Networks in Safety-Related Areas: Applications and Methods for Validation and Certification

CANCELED
Whistler, Canada, December 8th, 2001
 
Unfortunately, the workshop had to be canceled, because there were to few submissions.
If you are interested in this topic, please send us your email address so that we can contact you or even
set up a mailing list if there should be enough interest.
 
Workshop Organizers:
J. Schumann, RIACS/NASA Ames (schumann@email.arc.nasa.gov)
P. Lisboa, Liverpool John Moores University, (P.J.Lisboa@livjm.ac.uk)
R. Knaus, Instant Recall, Inc., (rknaus@irecall.com)

Over the recent years, Artificial Neural Networks have found their way into various safety-related and safety-critical areas, for example, power generation and transmission, transportation, avionics, environmental monitoring and control, medical applications, and consumer products. Applications range from classification to monitoring and control. Quite often, these applications proved to be highly successful, leading from pure research prototypes into serious experimental systems (e.g., a neural-network-based flight-control system test-flown on a NASA F-15ACTIVE) or commercial products (e.g., Sharp's Logi-cook).

However, the general question of how to make sure that the ANN-based system performs as expected in all cases has not yet been addressed satisfactorily. All safety-related software applications require careful verification and validation (V&V) of the software components, ranging from extended testing to full-fledged certification procedures (e.g., DO178-B). However, for neural-network based systems, a number of specific issues have to be addressed. For example, a lack of a concise plant model, often a major reason to use a ANN in the first place, makes traditional approaches to V&V impossible.

In this workshop, we will address such issues. In particular, we will discuss the following (non-exhaustive list of) topics:

  • theoretical methodologies to characterise the properties of neural network solutions, e.g., multiple realisations of a particular network and ways of managing this
  • fundamental software approaches to V&V and implications for ANNs, e.g., the application of FMEA
  • statistical (Bayesian) methods and symbolic techniques like rule extraction with subsequent V&V to assess and guarantee the performance of a ANN
  • dynamic monitoring of the ANN's behavior
  • stability proofs for control of dynamical systems with ANNs
  • principled approaches to design assurance, risk assessment, and performance evaluation of systems with ANNs
  • experience of application and certification of ANNs for safety-related applications
  • V&V techniques specifically suitable for on-line trained and adaptive systems


    • This workshop aims to bring together researchers who have applied ANNs in safety-related areas and actually addressed questions of demonstrating flawless operation of the ANN, researchers working on theoretical topics of convergence/performance assessment, researchers in the area of nonlinear adaptive control, and researchers from the area of formal methods in software design for safety-critical systems. Many prototypical/experimental application of neural networks in safety-related areas have demonstrated their usefulness successfully. But ANN applicability in safety-critical areas is substantially limited because of a lack of methods and techniques for verification and validation. Currently, there is no silver bullet for V&V in traditional software, and with the more complicated situation for ANNs, none is expected here in the short run. However, any result can have substantial impact in this field.
     

    Workshop format:

    Date: TBD (12/7 or 12/8 2001)

    This is a one-day workshop. It starts with an invited talk, describing a prominent neural-network based approach in a safety-critical domain (speaker TBD, perhaps a ANN based flight-control system) and discussing validation and certification issues. Two technical sessions with paper presentations and generous time for discussions will follow. The workshop will end with a (moderated) round-table discussion on important issues for future trends in this area. Specific focus will be laid upon the topic of how the traditional software design, implementation and test process needs to be adapted in order to be suited for neural-network based applications.
    Contributions: If you are planning to attend this workshop as a particpiant and/or are interested to present your work, please notify one of the workshop organizers no later than
    October 22, 2001.

    This email should contain name and tentative title of the presentation (if you are interested to give a presentation). A short (1-4 pages) abstract, technical paper, or position paper is due on
    October 29, 2001.

    Please send this abstract/paper as Postscript or PDF to schumann@email.arc.nasa.gov. Notification of acceptance will be sent out on November 5, 2001. All accepted contributions will be handed out to the participants as an informal workshop proceedings and will appear as a RIACS Technical Report.
     

    Registration and Travel: NIPS Registration and Travel


     
     
     
     
    Program Committee:
    A. Kelkar, Iowa State University, (akelkar@iastate.ed)
    R. Knaus, Instant Recall, Inc., (rknaus@irecall.com)
    P. Lisboa, Liverpool John Moores University, (P.J.Lisboa@livjm.ac.uk)
    A. Mili, West Virginia University, (amili@csee.wvu.edu)
    J. Schumann, RIACS/NASA Ames (schumann@email.arc.nasa.gov)

    This page: http://ase.arc.nasa.gov/people/schumann/workshops/NIPS2001 Questions? Send email to schumann@email.arc.nasa.gov
    $Id: index.html,v 1.7 2001/10/16 16:19:21 schumann Exp $