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

Digging deeper on "deep" learning: A computational ecology approach.

Massimo Buscema1, Pier Luigi Sacco2

  • 1Semeion Research Center,00128 Rome,Italy.m.buscema@semeion.it.www.semeion.itwww.researchgate.net/profile/Massimo_Buscema.

The Behavioral and Brain Sciences
|January 19, 2018
PubMed
Summary

This study introduces a novel artificial intelligence approach using diverse neural networks and associative memory, prioritizing adaptive learning over massive computation for more flexible AI. It emphasizes cross-domain learning inspired by human intelligence to advance AI development.

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

The asymmetric structure of male-female bonding explains gender differences in relationship dependence.

The Behavioral and brain sciences·2026
Same author

Significance is not enough: The biopsychological foundations of romantic love.

The Behavioral and brain sciences·2026
Same author

Cultural fiction as predictive-affective simulation: a neurocognitive model of male emotional regulation and gender violence prevention.

Frontiers in psychology·2026
Same author

Completeness as a fitness landscape for cognitive architectures.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Music as a scientific metaphor for mind and brain.

Neuroscience and biobehavioral reviews·2026
Same author

The Museum as a Mindful Space: Reducing Visitors' Stress and Anxiety Levels Through the ASBA Protocol.

Behavioral sciences (Basel, Switzerland)·2026

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Cognitive Science

Background:

  • Current deep learning models rely heavily on massive computation and extensive datasets.
  • There is a need for AI systems that exhibit greater adaptive flexibility and model-based learning capabilities.
  • Human intelligence demonstrates effective cross-domain learning and generalization.

Purpose of the Study:

  • To propose an alternative paradigm to deep learning based on computational ecologies.
  • To explore the role of structurally diverse artificial neural networks and associative memory in AI.
  • To integrate principles of human intelligence, specifically cross-domain learning, into artificial intelligence.

Main Methods:

  • Developing computational ecologies comprising structurally diverse artificial neural networks.

Related Experiment Videos

  • Investigating dynamic associative memory responses to various stimuli.
  • Implementing model-based learning and adaptive flexibility strategies.
  • Facilitating cross-fertilization of learning processes across multiple domains.

    • Main Results:

    • Demonstrated a shift from massive computation to model-based, adaptive learning.
    • Showcased the potential of diverse neural network architectures and associative memory.
    • Highlighted the efficacy of cross-domain learning for enhanced AI.
    • Proposed a framework for more flexible and human-like artificial intelligence.

    Conclusions:

    • An alternative approach to deep learning is presented, focusing on computational ecologies and adaptive flexibility.
    • The proposed method leverages diverse neural networks and associative memory for more robust AI.
    • Integrating cross-domain learning principles from human intelligence is crucial for advancing artificial intelligence.