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 Concept Videos

Survival Tree01:19

Survival Tree

172
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a...
172
Introduction to Learning01:18

Introduction to Learning

570
Learning is the process of acquiring knowledge or skills through practice or experience, leading to long-lasting behavioral changes. This acquisition occurs through interaction with the environment and requires practice or experience. For instance, mastering a skill such as surfing requires considerable practice and experience, highlighting the essential role of repeated interactions with the environment in learning.
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...
570
Observational Learning01:12

Observational Learning

348
Albert Bandura's observational learning, also known as imitation or modeling, occurs when a person observes and imitates another's behavior. It is a quicker process than operant conditioning. A well-known example is the Bobo doll study, where children who saw an adult acting aggressively towards the doll were more likely to act aggressively when left alone, compared to those who observed a nonaggressive adult. Many psychologists view observational learning as a form of latent learning...
348

You might also read

Related Articles

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

Sort by
Same author

Design and effectiveness evaluation of a web-based microplatform for mindfulness intervention for organ donor family members.

Internet interventions·2026
Same author

Direct observation of anomalous CO<sub>2</sub> dispersion in multi-scale porous media.

Lab on a chip·2026
Same author

Construct a digital psychological intervention program for organ donor families: An Intervention mapping study.

Digital health·2026
Same author

Decoupling of Neogene seawater lithium isotopes from uplift-driven weathering.

Nature communications·2026
Same author

Microscopic magnetic-field imaging of a single lunar dust grain.

Fundamental research·2026
Same author

Downregulation of NOX2 Expression Alleviates Severe Acute Pancreatitis by Restoring Autophagy and Inhibiting mtDNA-mediated Activation of the cGAS-STING Signalling Pathway.

Inflammation·2026
Same journal

AI-driven neuroanalytic modeling for mental health: multichannel CNN-based autism spectrum disorder detection via facial pattern analysis.

Frontiers in computational neuroscience·2026
Same journal

Modeling multiscale neural dynamics for EEG-based emotion recognition using an attentive wavelet-transformer framework.

Frontiers in computational neuroscience·2026
Same journal

New directions for complex systems in contemporary neuroscience: a morphodynamic and emergent function approach.

Frontiers in computational neuroscience·2026
Same journal

NMDA receptor kinetics drive distinct routes to chaotic firing in pyramidal neurons.

Frontiers in computational neuroscience·2026
Same journal

Schumann-anchored golden ratio organization of human neural oscillations.

Frontiers in computational neuroscience·2026
Same journal

Toward model-guided electrophysiology-Encoding of chirps in the electrosensory periphery of <i>Apteronotus leptorhynchus</i>.

Frontiers in computational neuroscience·2026
See all related articles

Related Experiment Video

Updated: Sep 27, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

662

Biologically Plausible Training Mechanisms for Self-Supervised Learning in Deep Networks.

Mufeng Tang1, Yibo Yang1, Yali Amit1

  • 1Department of Statistics, University of Chicago, Chicago, IL, United States.

Frontiers in Computational Neuroscience
|April 7, 2022
PubMed
Summary
This summary is machine-generated.

We developed biologically plausible training methods for self-supervised learning (SSL) in deep networks, using simpler computations and local learning rules. These methods achieve performance comparable to standard backpropagation for downstream tasks.

Keywords:
back-propagation (BP)difference target propagationhinge losslayerwise learningself-supervised learning

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
09:34

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

Published on: September 25, 2021

4.1K

Related Experiment Videos

Last Updated: Sep 27, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

662
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
09:34

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

Published on: September 25, 2021

4.1K

Area of Science:

  • Computational Neuroscience
  • Deep Learning Architectures
  • Machine Learning Algorithms

Background:

  • Self-supervised learning (SSL) in deep networks typically relies on complex computations and biologically implausible mechanisms.
  • Existing contrastive losses in SSL often involve intricate calculations like normalization and inner products, hindering biological plausibility.
  • Backpropagation, a standard training algorithm, presents challenges regarding biological plausibility due to its reliance on symmetric weights and complex error propagation.

Purpose of the Study:

  • To develop biologically plausible training mechanisms for self-supervised learning (SSL) in deep neural networks.
  • To propose a contrastive hinge-based loss function that avoids complex computations.
  • To introduce biologically plausible alternatives to backpropagation, such as difference target propagation (DTP) and layer-wise learning (GLL/RLL).

Main Methods:

  • Proposed a contrastive hinge-based loss function with simple, local computations.
  • Introduced difference target propagation (DTP) using target-based local losses and a Hebbian learning rule.
  • Implemented layer-wise learning (GLL/RLL) with modified backpropagation using fixed or updated random feedback weights.
  • Trained convolutional neural networks (CNNs) using SSL with the proposed DTP, GLL, and RLL methods.

Main Results:

  • The proposed contrastive hinge loss simplifies computations compared to standard contrastive losses.
  • DTP and layer-wise learning methods offer biologically plausible alternatives to backpropagation, addressing the symmetric weight issue.
  • CNNs trained with SSL and the proposed methods achieved performance comparable to standard backpropagation on downstream linear classifier evaluations.

Conclusions:

  • Biologically plausible training mechanisms can be effectively implemented for self-supervised learning in deep networks.
  • The proposed contrastive hinge loss and alternative propagation methods offer practical and plausible approaches for neural network training.
  • These findings contribute to bridging the gap between artificial neural networks and biological neural systems in terms of learning mechanisms.