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

Sampling Continuous Time Signal01:11

Sampling Continuous Time Signal

218
In signal processing, a continuous-time signal can be sampled using an impulse-train sampling technique, followed by the zero-order hold method. Impulse-train sampling involves the use of a periodic impulse train, which consists of a series of delta functions spaced at regular intervals determined by the sampling period. When a continuous-time signal is multiplied by this impulse train, it generates impulses with amplitudes corresponding to the signal's values at the sampling points.
In the...
218
Introduction to Learning01:18

Introduction to Learning

342
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...
342
Observational Learning01:12

Observational Learning

149
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...
149
Streamlines, Streaklines, and Pathlines01:18

Streamlines, Streaklines, and Pathlines

1.1K
A streamline represents the trajectory that is always tangent to the fluid's velocity vector at any given point. The velocity of a fluid particle is always directed along the streamline, ensuring the particle continuously follows the streamline's path. Streamlines are particularly useful for visualizing the overall direction of flow in a fluid system, and they provide an instantaneous representation of the flow's velocity field. In steady flow, where conditions do not change over...
1.1K
Cognitive Learning01:21

Cognitive Learning

229
Cognitive learning is based on purposive behavior, incidental learning, and insight learning.
E. C. Tolman's theory of purposive behavior emphasizes that much behavior is goal-directed. He argued that to understand behavior, we must look at the entire sequence of actions leading to a goal. For instance, high school students study hard, not just due to past reinforcement but also to achieve the goal of getting into a good college.
Tolman introduced the idea that behavior is influenced by...
229
Stream Function01:20

Stream Function

1.3K
In two-dimensional incompressible fluid flow, the continuity equation is essential for ensuring mass conservation, meaning that any change in fluid entering or exiting a region is balanced by a corresponding change elsewhere. For incompressible flow, where density remains constant, this requirement simplifies to the condition that the divergence of the velocity field must be zero. Mathematically, this is expressed as,
1.3K

You might also read

Related Articles

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

Sort by
Same author

Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Holistic Invariant Retracing for Distortion-Resilient Multi-Modal Learning in Spatial Transcriptomics.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Demonstration of efficient predictive surrogates for large-scale quantum processors.

Nature communications·2026
Same author

A DeepSeek-powered AI system for automated chest radiograph interpretation in clinical practice.

Nature communications·2026
Same author

NoisePO: Efficient Semantic Noise Generation and Ranking for Diffusion-Based Text-to-Image Synthesis.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Stability and Generalization for Distributed SGDA.

IEEE transactions on pattern analysis and machine intelligence·2026

Related Experiment Video

Updated: Jun 12, 2025

A Multilayer Microfluidic Platform for the Conduction of Prolonged Cell-Free Gene Expression
11:23

A Multilayer Microfluidic Platform for the Conduction of Prolonged Cell-Free Gene Expression

Published on: October 6, 2019

10.2K

Continual Learning From a Stream of APIs.

Enneng Yang, Zhenyi Wang, Li Shen

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |September 20, 2024
    PubMed
    Summary

    This study introduces a new framework for continual learning (CL) using APIs, addressing data scarcity by generating pseudo-data. The method effectively distills knowledge from API streams, mitigating catastrophic forgetting in data-efficient and data-free scenarios.

    More Related Videos

    Continuous Hydrologic and Water Quality Monitoring of Vernal Ponds
    06:37

    Continuous Hydrologic and Water Quality Monitoring of Vernal Ponds

    Published on: November 13, 2017

    9.2K
    Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
    11:54

    Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

    Published on: May 8, 2021

    4.3K

    Related Experiment Videos

    Last Updated: Jun 12, 2025

    A Multilayer Microfluidic Platform for the Conduction of Prolonged Cell-Free Gene Expression
    11:23

    A Multilayer Microfluidic Platform for the Conduction of Prolonged Cell-Free Gene Expression

    Published on: October 6, 2019

    10.2K
    Continuous Hydrologic and Water Quality Monitoring of Vernal Ponds
    06:37

    Continuous Hydrologic and Water Quality Monitoring of Vernal Ponds

    Published on: November 13, 2017

    9.2K
    Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
    11:54

    Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

    Published on: May 8, 2021

    4.3K

    Area of Science:

    • Machine Learning
    • Artificial Intelligence
    • Continual Learning

    Background:

    • Continual learning (CL) aims to enable models to learn new tasks without forgetting previous ones.
    • Traditional CL methods require large datasets, which are often inaccessible due to privacy and copyright issues.
    • Machine Learning as a Service (MLaaS) via APIs offers an alternative, but presents new challenges for CL.

    Purpose of the Study:

    • To propose novel data-efficient CL (DECL-APIs) and data-free CL (DFCL-APIs) settings utilizing API streams.
    • To develop a data-free cooperative continual distillation learning framework to address challenges like unknown model parameters and catastrophic forgetting.

    Main Methods:

    • A cooperative distillation framework with two generators and one CL model trained adversarially.
    • Pseudo-data generation by querying APIs to distill knowledge into the CL model.
    • A network similarity regularization term to prevent forgetting of previous APIs.

    Main Results:

    • The proposed method achieves performance comparable to classic CL with full data in the DFCL-APIs setting on MNIST and SVHN.
    • In the DECL-APIs setting, the method reaches 0.97x, 0.75x, and 0.69x of classic CL performance on CIFAR10, CIFAR100, and MiniImageNet, respectively.

    Conclusions:

    • The proposed data-free cooperative continual distillation learning framework effectively addresses data scarcity in CL using APIs.
    • The method demonstrates strong performance in both data-efficient and data-free continual learning scenarios, outperforming traditional approaches in challenging settings.