Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Processing images by semi-linear predictability minimization.

N N Schraudolph1, M Eldracher, J Schmidhuber

  • 1IDSIA, Lugano, Switzerland. nic@idsia.ch

Network (Bristol, England)
|June 23, 1999
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

African swine fever: a global factor affecting agricultural markets over the medium term.

Revue scientifique et technique (International Office of Epizootics)·2022
Same author

Optimising traceability in trade for live animals and animal products with digital technologies.

Revue scientifique et technique (International Office of Epizootics)·2020
Same author

Salvage therapy with everolimus reduces the severity of treatment-refractory chronic GVHD without impairing disease control: a dual center retrospective analysis.

Bone marrow transplantation·2014
Same author

Introductory lecture the epidemiology and determinants of obesity in developed and developing countries.

International journal for vitamin and nutrition research. Internationale Zeitschrift fur Vitamin- und Ernahrungsforschung. Journal international de vitaminologie et de nutrition·2007
Same author

Learning nonregular languages: a comparison of simple recurrent networks and LSTM.

Neural computation·2002
Same author

Learning to forget: continual prediction with LSTM.

Neural computation·2000

Predictability minimization trains systems to develop visual feature detectors by making them unpredictable. This approach co-evolves feature detectors and predictors, leading to the emergence of known visual processing capabilities.

Area of Science:

  • Computational neuroscience
  • Artificial intelligence
  • Computer vision

Background:

  • The predictability minimization approach involves adaptive feature detectors and predictors.
  • Feature detectors aim for unpredictability, while predictors attempt to forecast their outputs.
  • This creates a co-evolutionary dynamic between detectors and predictors.

Purpose of the Study:

  • To implement and evaluate a visual processing system using semi-linear predictability minimization.
  • To investigate the system's response to diverse image datasets.
  • To determine if predictability minimization leads to the development of established visual feature detectors.

Main Methods:

  • Implementation of a visual processing system based on predictability minimization.

Related Experiment Videos

  • Training the system using semi-linear predictability minimization.
  • Experimentation with artificial and real-world image inputs.

    • Main Results:

    • The system successfully developed well-known visual feature detectors.
    • Predictability minimization consistently resulted in the emergence of these detectors across various conditions.
    • The co-evolution of feature detectors and predictors was observed.

    Conclusions:

    • Predictability minimization is an effective method for training visual processing systems.
    • This approach can autonomously develop specialized visual feature detectors.
    • The findings support the role of unpredictability in the development of complex sensory processing systems.