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

Activation and Inactivation of G Proteins01:22

Activation and Inactivation of G Proteins

8.7K
Heterotrimeric G proteins are guanine nucleotide-binding proteins. As the name suggests, heterotrimeric G proteins are composed of three subunits: alpha, beta, and gamma. They remain GDP-bound or GTP-bound inside the cells and switch between inactive/active states. The Gα subunit possesses the nucleotide-binding pocket that binds guanine nucleotides and switches between GDP or GTP-bound states. In contrast, the Gꞵ and Gγ subunits are always bound together with high...
8.7K
Types of Functions I01:26

Types of Functions I

80
Functions are fundamental mathematical tools that capture relationships between variables and describe how one quantity changes in relation to another. Their diverse forms allow them to model various real-world phenomena with precision and flexibility. Among the various categories, algebraic functions are prominent due to their formulation through basic arithmetic operations: addition, subtraction, multiplication, division, and root extraction.Algebraic functions include polynomial, rational,...
80
Neural Regulation01:37

Neural Regulation

41.6K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
41.6K
Types of Functions II01:19

Types of Functions II

61
Trigonometric and exponential functions are essential mathematical tools used to model distinct types of real-world behavior, particularly in periodic and growth-related phenomena. These functions extend the capabilities of basic algebraic models by capturing recurring cycles and rapid changes across various scientific and engineering contexts.Trigonometric functions, such as sine and cosine, are particularly effective for representing periodic phenomena. Their cyclic behavior makes them...
61
Activation Energy01:26

Activation Energy

85.0K
Activation energy is the minimum amount of energy necessary for a chemical reaction to move forward. The higher the activation energy, the slower the rate of the reaction. However, adding heat to the reaction will increase the rate, since it causes molecules to move faster and increase the likelihood that molecules will collide. The collision and breaking of bonds represents the uphill phase of a reaction and generates the transition state. The transition state is an unstable high-energy state...
85.0K
Types of Functions III01:28

Types of Functions III

75
Logarithmic and piecewise functions play central roles in mathematical modeling, particularly when capturing nonlinear or segmented behaviors in real-world phenomena. Although these functions differ fundamentally in structure and application, both serve to represent complex relationships in simplified mathematical terms.A logarithmic function is defined as the inverse of an exponential function, expressed as These functions grow quickly for small values of x but slow down as x increases,...
75

You might also read

Related Articles

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

Sort by
Same author

Point-of-Care Ocular Ultrasound as an Adjunctive Tool for Suspected Intracranial Hypertension in a Child.

Pediatrics international : official journal of the Japan Pediatric Society·2026
Same author

False but phonologically plausible linguistic priors induce cross-linguistic auditory illusions and attenuate electrophysiological markers of surprise.

Imaging neuroscience (Cambridge, Mass.)·2026
Same author

A predictive processing framework for joint action and communication.

Physics of life reviews·2026
Same author

Enhancing pix2pix with Swin Transformer for Cross Modal Brain CT-MR synthesis.

Research square·2025
Same author

Integrating large language models and active inference to understand eye movements in reading and dyslexia.

Physics of life reviews·2025
Same author

Unveiling key pathomic features for automated diagnosis and Gleason grade estimation in prostate cancer.

BMC medical imaging·2025

Related Experiment Video

Updated: Nov 16, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.1K

A survey on modern trainable activation functions.

Andrea Apicella1, Francesco Donnarumma2, Francesco Isgrò1

  • 1Dipartimento di Ingegneria Elettrica e delle Tecnologie dell'Informazione, Università di Napoli Federico II, Italy.

Neural Networks : the Official Journal of the International Neural Network Society
|February 21, 2021
PubMed
Summary
This summary is machine-generated.

Researchers surveyed trainable activation functions for neural networks, finding many models are equivalent to fixed functions with constrained weights. This research clarifies adaptable activation function approaches for improved network performance.

Keywords:
Activation functionsLearnable activation functionsMachine learningNeural networksTrainable activation functions

More Related Videos

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions
07:34

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

Published on: March 25, 2014

10.1K

Related Experiment Videos

Last Updated: Nov 16, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.1K
A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions
07:34

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

Published on: March 25, 2014

10.1K

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Deep Learning

Background:

  • Neural network performance is significantly influenced by the choice of activation functions.
  • There is a growing interest in trainable activation functions that adapt during the learning process, potentially enhancing network efficacy.

Purpose of the Study:

  • To survey and categorize diverse models of trainable activation functions.
  • To analyze the properties, advantages, and limitations of adaptable activation function approaches.
  • To clarify the relationship between trainable activation functions and fixed activation functions with constrained weights.

Main Methods:

  • Literature review and analysis of existing trainable activation function models.
  • Development of a taxonomy for classifying trainable activation functions.
  • Comparative analysis of different trainable activation function approaches.

Main Results:

  • Identified a wide variety of trainable activation function models.
  • Proposed a taxonomy to organize these diverse models.
  • Demonstrated that many trainable activation functions are mathematically equivalent to fixed activation functions combined with specific weight constraints.

Conclusions:

  • Trainable activation functions offer potential performance improvements in neural networks.
  • A clear taxonomy helps understand the landscape of adaptable activation functions.
  • Understanding the equivalencies to fixed functions with constraints is crucial for practical implementation and theoretical analysis.