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

The concave-convex procedure.

A L Yuille1, Anand Rangarajan

  • 1Smith-Kettlewell Eye Research Institute, San Francisco, CA 94115, USA. yuille@ski.org

Neural Computation
|April 12, 2003
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

Toward Improving the Generation Quality of Autoregressive Slot VAEs.

Neural computation·2024
Same author

Learning Canonical Embeddings for Unsupervised Shape Correspondence With Locally Linear Transformations.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

Machine-Learning-Based Real-Time Multi-Camera Vehicle Tracking and Travel-Time Estimation.

Journal of imaging·2022
Same author

Spherical Minimum Description Length.

Entropy (Basel, Switzerland)·2020
Same author

Erratum to "Annotated normal CT data of the abdomen for deep learning: Challenges and strategies for implementation" [Diagn. Interv. Imaging. 101 (2020) 35-44].

Diagnostic and interventional imaging·2020
Same author

Differentiating autoimmune pancreatitis from pancreatic ductal adenocarcinoma with CT radiomics features.

Diagnostic and interventional imaging·2020
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

The concave-convex procedure (CCCP) offers a unified framework for designing optimization algorithms. This method guarantees monotonic decrease in global optimization and energy functions, applicable to diverse problems.

Area of Science:

  • Optimization theory
  • Dynamical systems
  • Machine learning algorithms

Background:

  • Many optimization algorithms lack a unified theoretical framework.
  • Understanding algorithm convergence and generating new methods remains a challenge.

Purpose of the Study:

  • To introduce the concave-convex procedure (CCCP) as a general method for constructing discrete-time iterative dynamical systems.
  • To demonstrate the broad applicability of CCCP across various optimization problems and existing algorithms.

Main Methods:

  • Developing the concave-convex procedure (CCCP) for iterative dynamical systems.
  • Re-expressing existing algorithms, including Expectation-Maximization (EM), Legendre minimization, variational bounding, neural network, mean-field theory, generalized iterative scaling, and Sinkhorn's algorithms, in terms of CCCP.

Related Experiment Videos

  • Proving monotonic decrease of global optimization and energy functions.

    • Main Results:

    • CCCP provides a unified approach to optimization algorithm design and analysis.
    • Expectation-Maximization (EM), Legendre minimization, variational bounding, neural network, mean-field theory, generalized iterative scaling, and Sinkhorn's algorithms are shown to be instances of CCCP.
    • The procedure guarantees monotonic decrease in optimization and energy functions.

    Conclusions:

    • CCCP serves as a powerful tool for understanding and proving the convergence of existing optimization algorithms.
    • CCCP facilitates the generation of novel optimization algorithms with guaranteed convergence properties.
    • The framework unifies diverse algorithmic approaches within a single, coherent methodology.