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

Aggregates Classification01:29

Aggregates Classification

741
Aggregate classification is generally based on its size, petrographic characteristics, weight, and source. Size classification ranges from coarse to fine aggregates, defined by the size of the particles. Coarse aggregates are particles that do not pass through ASTM sieve No. 4, and aggregates that pass through the sieve are fine aggregates.
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
741
Associative Learning01:27

Associative Learning

1.0K
Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...
1.0K
Collisions in Multiple Dimensions: Problem Solving01:06

Collisions in Multiple Dimensions: Problem Solving

5.0K
In multiple dimensions, the conservation of momentum applies in each direction independently. Hence, to solve collisions in multiple dimensions, we should write down the momentum conservation in each direction separately. To help understand collisions in multiple dimensions, consider an example.
A small car of mass 1,200 kg traveling east at 60 km/h collides at an intersection with a truck of mass 3,000 kg traveling due north at 40 km/h. The two vehicles are locked together. What is the...
5.0K
Survival Tree01:19

Survival Tree

312
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...
312
Masking and Demasking Agents01:19

Masking and Demasking Agents

3.3K
EDTA titrations may necessitate masking and demasking agents to temporarily protect a particular metal ion in a mixture from the EDTA reaction. These agents facilitate the sequential analysis of the metal ions by forming stable complexes with some—but not all—metal ions during certain steps.
There are many masking agents, such as cyanide, fluoride, triethanolamine, thiourea, and 2,3-bis(sulfanyl)propan-1-ol (formerly 2,3-dimercapto-1-propanol), with the masking agent chosen based on...
3.3K
Aggression01:47

Aggression

30.0K
Humans engage in aggression when they seek to cause harm or pain to another person. Aggression takes two forms depending on one’s motives: hostile or instrumental. Hostile aggression is motivated by feelings of anger with intent to cause pain; a fight in a bar with a stranger is an example of hostile aggression. In contrast, instrumental aggression is motivated by achieving a goal and does not necessarily involve intent to cause pain (Berkowitz, 1993); a contract killer who murders for...
30.0K

You might also read

Related Articles

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

Sort by
Same author

Unsupervised Data Driven Clustering of the Neurological Assessments of People With Traumatic SCI Focusing on Sensorimotor Complete Injuries.

Neurorehabilitation and neural repair·2026
Same author

Linking neurological status to functional outcomes in spinal cord injury: a multi-class, task-specific approach.

BMC biomedical engineering·2026
Same author

MV-MR: Multi-Views and Multi-Representations for Self-Supervised Learning and Knowledge Distillation.

Entropy (Basel, Switzerland)·2024
Same author

Hubble Meets Webb: Image-to-Image Translation in Astronomy.

Sensors (Basel, Switzerland)·2024
Same author

Differential privacy preserved federated learning for prognostic modeling in COVID-19 patients using large multi-institutional chest CT dataset.

Medical physics·2024
Same author

PRIMIS: Privacy-preserving medical image sharing via deep sparsifying transform learning with obfuscation.

Journal of biomedical informatics·2024

Related Experiment Videos

Machine learning through cryptographic glasses: combating adversarial attacks by key-based diversified aggregation.

Olga Taran1, Shideh Rezaeifar1, Taras Holotyak1

  • 1Stochastic Information Processing Group, Department of Computer Science, University of Geneva, 7 Route de Drize, Carouge, GE Switzerland.

EURASIP Journal on Information Security
|July 21, 2020
PubMed
Summary
This summary is machine-generated.

This study introduces a key-based diversified aggregation (KDA) to defend deep neural networks (DNN) against adversarial attacks. KDA uses a secret key for randomization, preventing gradient back-propagation and enhancing classifier robustness.

Keywords:
Adversarial examplesBlack-box attacksDefenseDiversified aggregationMachine learningNon-gradient/gradient-based attacksRandomization

Related Experiment Videos

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Cybersecurity

Background:

  • Deep neural networks (DNNs) are widely used but vulnerable to adversarial attacks.
  • This vulnerability limits their deployment in critical applications.
  • Understanding DNN vulnerability mechanisms and developing robust defenses are crucial.

Purpose of the Study:

  • To propose a novel defense mechanism, Key-Based Diversified Aggregation (KDA), against adversarial attacks on DNN classifiers.
  • To enhance the robustness and reliability of DNN models in gray- and black-box scenarios.
  • To explore the synergy between machine learning and cryptography for defense.

Main Methods:

  • Introduced Key-Based Diversified Aggregation (KDA) utilizing a secret key for randomization.
  • Implemented key-based randomization across multiple channels in a special transform domain.
  • Aggregated soft outputs from diversified channels to stabilize results and improve reliability.
  • Designed KDA to prevent gradient back-propagation and bypass attacks.

Main Results:

  • Experimental evaluation demonstrated high robustness and universality of KDA.
  • KDA effectively countered state-of-the-art gradient-based gray-box transferability attacks.
  • KDA proved effective against non-gradient-based black-box attacks.
  • The secret key provided a significant information advantage to the defender.

Conclusions:

  • KDA offers a robust and universal defense strategy against adversarial attacks on DNN classifiers.
  • The key-based randomization mechanism successfully prevents gradient back-propagation.
  • KDA enhances the reliability and security of DNNs for critical applications.