Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Survival Tree

Survival Tree

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...

Residuals and Least-Squares Property

Residuals and Least-Squares Property

The vertical distance between the actual value of y and the estimated value of y. In other words, it measures the vertical distance between the actual data point and the predicted point on the line
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...

Law of Independent Assortment

Law of Independent Assortment

While Mendel’s Law of Segregation states that the two alleles for one gene are separated into different gametes, a different question of how different genes are inherited remains. For example, is the gene for tall plants inherited with the gene for green peas? Mendel asked this question by experimenting with a dihybrid cross; a cross in which both parents are homozygous for two distinct traits resulting in an F1 generation that are heterozygous for both traits.

Overview Of Cell Separation And Isolation

Overview Of Cell Separation And Isolation

Cell separation was first achieved in 1964 by S. H. Seal, who separated large tumor cells from the smaller blood cells using filtration. Two years later, Pohl and Hawk performed experiments on how cells respond differently to a nonuniform electric field based on the cell type. Such observations were the inception of cell separation methods, which allow isolating a single cell type from a heterogeneous sample.

Law of Segregation

Law of Segregation

When crossing pea plants, Mendel noticed that one of the parental traits would sometimes disappear in the first generation of offspring, called the F1 generation, and could reappear in the next generation (F2). He concluded that one of the traits must be dominant over the other, thereby causing masking of one trait in the F1 generation. When he crossed the F1 plants, he found that 75% of the offspring in the F2 generation had the dominant phenotype, while 25% had the recessive phenotype.

Extraction: Partition and Distribution Coefficients

Extraction: Partition and Distribution Coefficients

The distribution law or Nernst's distribution law is the law that governs the distribution of a solute between two immiscible solvents. This law, also known as the partition law, states that if a solute is added to the mixture of two immiscible solvents at a constant temperature, the solute is distributed between the two solvents in such a way that the ratio of solute concentrations in the solvents remains constant at equilibrium.
For extracting a solute from an aqueous phase into an...

You might also read

Related Articles

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

Sort by

Same author

The 2020 US Decennial Census is more private than you (might) think.

Proceedings of the National Academy of Sciences of the United States of America·2025

Same author

Exploring deep neural networks via layer-peeled model: Minority collapse in imbalanced training.

Proceedings of the National Academy of Sciences of the United States of America·2021

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 17, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

A law of data separation in deep learning.

Hangfeng He^1,2, Weijie J Su³

¹Department of Computer Science, University of Rochester, Rochester, NY 14627.

Proceedings of the National Academy of Sciences of the United States of America

|August 28, 2023

Summary

This summary is machine-generated.

Researchers discovered a quantitative law explaining how deep neural networks (DNNs) separate data by class in intermediate layers. This finding offers insights into DNNs, aiding AI design and interpretation.

Keywords:

constant geometric rate data separation deep learning intermediate layers

More Related Videos

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

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

Published on: December 15, 2023

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Related Experiment Videos

Last Updated: Jul 17, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

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

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

Published on: December 15, 2023

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Area of Science:

Artificial Intelligence
Machine Learning
Computer Science

Background:

Deep learning models have advanced AI but their 'black-box' nature complicates design and interpretation.
Understanding intermediate layer data processing is crucial for AI development and high-stakes applications.

Purpose of the Study:

To investigate how deep neural networks (DNNs) process data within their intermediate layers.
To uncover a fundamental, quantitative law governing data separation in DNNs.

Main Methods:

Analysis of data separation within intermediate layers of various deep neural network architectures.
Observation of the law's emergence during the training process across diverse datasets.

Main Results:

A simple, quantitative law was identified that governs how DNNs separate data by class across all layers.
Each layer consistently improves data separation at a constant geometric rate.
This law was observed across multiple network architectures and datasets during training.

Conclusions:

The discovered law provides a fundamental understanding of DNN data processing.
It offers practical guidelines for designing better AI architectures.
The findings can enhance model robustness, out-of-sample performance, and prediction interpretability.