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

Energy Associated With a Charge Distribution01:21

Energy Associated With a Charge Distribution

1.5K
The work done to bring a charge through a distance r is given by the potential difference between the initial and the final position. To assemble a collection of point charges, the total work done can be expressed in terms of the product of each pair of charges divided by their separation distance, defined with respect to a suitable origin. Solving this expression gives the energy stored in a point charge distribution.
1.5K
Circuit Terminology01:14

Circuit Terminology

636
An electrical network is a system composed of interconnected elements, such as resistors, capacitors, inductors, and voltage or current sources. Unlike a circuit, an electrical network does not necessarily form a closed path. In other words, while all circuits can be considered networks due to their interconnected nature, not every network qualifies as a circuit.
A circuit, on the other hand, is also an interconnected system of electrical elements but must contain one or more closed paths.
636
Entropy Change in Reversible Processes01:10

Entropy Change in Reversible Processes

2.5K
In the Carnot engine, which achieves the maximum efficiency between two reservoirs of fixed temperatures, the total change in entropy is zero. The observation can be generalized by considering any reversible cyclic process consisting of many Carnot cycles. Thus, it can be stated that the total entropy change of any ideal reversible cycle is zero.
The statement can be further generalized to prove that entropy is a state function. Take a cyclic process between any two points on a p-V diagram.
2.5K
Energy Losses in Transformers01:21

Energy Losses in Transformers

862
In an ideal transformer, it is assumed that there are no energy losses, and, hence, all the power at the primary winding is transferred to the secondary winding. However, in reality,  the transformers always have some energy losses, and, hence, the output power obtained at the secondary winding is less than the input power at the primary winding due to energy losses.
There are four main reasons for energy losses in transformers.
The first cause can be  the high resistance of the...
862
Equivalent Capacitance01:19

Equivalent Capacitance

333
From the study of resistive circuits, it is understood that employing a series-parallel combination serves as an effective strategy for simplifying circuits. Capacitors can be arranged within a circuit in one of two ways: a series configuration or a parallel configuration. The way these capacitors are connected to a battery will influence both the potential drop across each individual capacitor and the size of the charge that each capacitor can store. This is determined by the specific type of...
333
Complex Power01:14

Complex Power

402
Power engineers have introduced the concept of complex power to determine the cumulative effect of parallel loads. This idea plays a crucial role in power analysis because it encompasses all the details related to the power consumed by a specific load.
Complex power is defined as the multiplication of the voltage and the complex conjugate of the current. The magnitude of this power, known as apparent power, is measured in volt-amperes (VA). Notably, the angle of the complex power equates to the...
402

You might also read

Related Articles

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

Sort by
Same author

Energy Complexity of Convolutional Neural Networks.

Neural computation·2024
Same author

Rotation-based schedules in elementary schools to prevent COVID-19 spread: a simulation study.

Scientific reports·2023
Same author

On the contact tracing for COVID-19: A simulation study.

Epidemics·2023
Same author

Delays, Masks, the Elderly, and Schools: First Covid-19 Wave in the Czech Republic.

Bulletin of mathematical biology·2022
Same author

Importance of vaccine action and availability and epidemic severity for delaying the second vaccine dose.

Scientific reports·2022
Same author

Effective Automatic Method Selection for Nonlinear Regression Modeling.

International journal of neural systems·2021

Related Experiment Video

Updated: Jun 24, 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

520

On energy complexity of fully-connected layers.

Jiří Šíma1, Jérémie Cabessa2, Petra Vidnerová1

  • 1Institute of Computer Science of the Czech Academy of Sciences, Pod Vodárenskou věží 271/2, Prague 8, 182 00, Czechia.

Neural Networks : the Official Journal of the International Neural Network Society
|June 11, 2024
PubMed
Summary

Deep neural networks (DNNs) consume significant energy. This study analyzes energy complexity for fully-connected layers, establishing an optimal quadratic bound for efficient hardware deployment in mobile devices.

Keywords:
Convolutional neural networksDataflowDeep neural networksEnergy complexityEnergy consumptionFully-connected layer

More Related Videos

Microstate and Omega Complexity Analyses of the Resting-state Electroencephalography
06:40

Microstate and Omega Complexity Analyses of the Resting-state Electroencephalography

Published on: June 15, 2018

10.2K
Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.0K

Related Experiment Videos

Last Updated: Jun 24, 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

520
Microstate and Omega Complexity Analyses of the Resting-state Electroencephalography
06:40

Microstate and Omega Complexity Analyses of the Resting-state Electroencephalography

Published on: June 15, 2018

10.2K
Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.0K

Area of Science:

  • Computer Science
  • Electrical Engineering
  • Artificial Intelligence

Background:

  • Deep neural networks (DNNs) face increasing energy demands due to their growing size.
  • Hardware implementations of DNNs are critical for deployment on low-power mobile devices.
  • Previous work established a validated, hardware-independent energy complexity model for convolutional neural networks (CNNs).

Purpose of the Study:

  • To theoretically analyze the energy complexity of fully-connected layers in DNNs.
  • To determine the optimal energy complexity considering memory transfers between DRAM and Buffer.
  • To validate the findings on Simba and Eyeriss hardware.

Main Methods:

  • Theoretical analysis of energy complexity for fully-connected layer computation.
  • Establishing a general lower bound on energy complexity.
  • Presenting two dataflows to derive upper bounds on energy costs.
  • Utilizing linear programming's weak duality theorem for optimization.
  • Experimental validation on Simba and Eyeriss hardware.

Main Results:

  • A general lower bound for the energy complexity of fully-connected layers was established.
  • Two dataflows were analyzed, yielding upper bounds for energy costs.
  • Optimal quadratic energy complexity was proven for partitioned Buffer memory.
  • Experimental validation confirmed the asymptotically optimal quadratic energy complexity.

Conclusions:

  • The study establishes the optimal quadratic energy complexity for fully-connected layers.
  • This finding is crucial for deploying large DNNs efficiently on low-power hardware.
  • The results contribute to the development of energy-efficient AI hardware.