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

A recurrent neural network for nonlinear optimization with a continuously differentiable objective function and bound

X B Liang1, J Wang

  • 1Department of Electrical and Computer Engineering, University of Delaware, Newark, DE 19716, USA.

IEEE Transactions on Neural Networks
|February 6, 2008
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

Reaction of cytochrome bo3 with oxygen: extra redox center(s) are present in the protein.

Biochemistry·1995
Same author

Regulation of striatal cyclic-3',5'-adenosine monophosphate accumulation and GABA release by glutamate metabotropic and dopamine D1 receptors.

The Journal of pharmacology and experimental therapeutics·1995
Same author

Calculation of relative binding free energies and configurational entropies: a structural and thermodynamic analysis of the nature of non-polar binding of thrombin inhibitors based on hirudin55-65.

Journal of molecular biology·1995
Same author

Hydroxylated aromatic inhibitors of HIV-1 integrase.

Journal of medicinal chemistry·1995
Same author

Sevoflurane for outpatient anesthesia: a comparison with propofol.

Anesthesia and analgesia·1995
Same author

Resident bone marrow macrophages produce type 1 interferons that can selectively inhibit interleukin-7-driven growth of B lineage cells.

Immunity·1995
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces a continuous-time recurrent neural network for solving nonlinear optimization problems with bound constraints. The model demonstrates convergence and stability, making it effective for various optimization tasks.

Area of Science:

  • Computational Mathematics
  • Artificial Intelligence
  • Optimization Theory

Background:

  • Nonlinear optimization problems with bound constraints are prevalent in various scientific and engineering fields.
  • Existing methods may face challenges with convergence and stability for complex objective functions.
  • Recurrent neural networks offer a potential framework for addressing these optimization challenges.

Purpose of the Study:

  • To propose a novel continuous-time recurrent neural-network model for solving nonlinear optimization problems with bound constraints.
  • To analyze the theoretical properties of the proposed network, including regularity, completeness, and convergence.
  • To demonstrate the model's effectiveness through simulations for nonlinear and quadratic optimization tasks.

Main Methods:

Related Experiment Videos

  • Development of a continuous-time recurrent neural network architecture.
  • Mathematical analysis of network properties: regularity, completeness, primal-dual convergence, and attractivity.
  • Simulation studies to evaluate performance on nonlinear and strictly convex quadratic optimization problems with bound constraints.

Main Results:

  • The recurrent neural network model ensures that optima correspond to network equilibria, demonstrating regularity.
  • For convex objective functions, the network exhibits completeness, with equilibria matching the function's optima.
  • The network shows primal-dual convergence within the feasible region and attractivity from outside, ensuring stable convergence.
  • Global exponential stability is proven for minimizing strictly convex quadratic functions under specific network parameters.

Conclusions:

  • The proposed continuous-time recurrent neural network is a robust tool for nonlinear optimization with bound constraints.
  • The model's theoretical properties guarantee convergence and stability for a wide range of optimization problems.
  • Simulation results validate the network's practical performance and potential for real-world applications.