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 Experiment Videos

Segmented-memory recurrent neural networks.

Jinmiao Chen1, Narendra S Chaudhari

  • 1School of Computer Engineering, Nanyang Technological University, Singapore 639798, Singapore. chenjinmiao@pmail.ntu.edu.sg

IEEE Transactions on Neural Networks
|July 17, 2009
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Neural Circuits01:25

Neural Circuits

Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Chunking and Rehearsal in Sensory Memory01:22

Chunking and Rehearsal in Sensory Memory

Improving short-term memory can be achieved through techniques like chunking and rehearsal. Chunking involves organizing information into larger, more manageable units. This technique is particularly useful for information that exceeds the typical memory span of between five and nine items. For instance, logging into an online account with a password like "ta89vq0179gz" involves grouping letters and numbers into three chunks—ta89, vq01, and 79gz. It makes large amounts of information more...

You might also read

Related Articles

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

Sort by
Same author

Direct contact between iPSC-derived macrophages and hepatocytes drives reciprocal acquisition of Kupffer cell identity and hepatocyte maturation.

eLife·2026
Same author

Ketogenic Diet Alleviates Colorectal Cancer by Attenuating Macrophage M2 Polarisation Triggered by Oncometabolite MMA Derived From the Gut Microbiota.

Cell proliferation·2026
Same author

AI-guided prediction of ncRNA biochemical features for therapeutic targeting.

Trends in pharmacological sciences·2026
Same author

Mosaic integration of spatial multi-omics with SpaMosaic.

Nature genetics·2026
Same author

YOLO-SDA: an innovative YOLOv12-derived model with superior performance in recognizing peanut foliar diseases.

Frontiers in plant science·2026
Same author

Molecular subtyping and prognostic modeling of non-small cell lung cancer based on disulfidptosis-related genes.

Translational cancer research·2026
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Segmented-memory recurrent neural networks (SMRNNs) improve learning long-term dependencies by processing data in segments. This novel architecture outperforms conventional recurrent neural networks (RNNs) on complex sequence prediction tasks.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Conventional recurrent neural networks (RNNs) struggle with learning long-term dependencies in sequential data.
  • This limitation hinders performance in tasks requiring memory of distant past events.

Purpose of the Study:

  • To introduce a novel architecture, the segmented-memory recurrent neural network (SMRNN), designed to overcome RNN limitations.
  • To evaluate the SMRNN's effectiveness in learning long-term temporal dependencies.

Main Methods:

  • The SMRNN architecture breaks symbolic sequences into segments, processing them with separate symbol-level and segment-level contexts.
  • An extended real-time recurrent learning algorithm was employed for SMRNN training.
  • Performance was evaluated on information latching, the two-sequence problem, and protein secondary structure (PSS) prediction.

Related Experiment Videos

Main Results:

  • SMRNN demonstrated superior performance on long-term dependency problems compared to conventional RNNs.
  • Theoretical analysis confirmed the benefit of segmented memory for learning long-term temporal dependencies.
  • The impact of segment length on performance was investigated.

Conclusions:

  • The SMRNN architecture effectively addresses the challenge of learning long-term dependencies in sequential data.
  • SMRNN offers a promising alternative to conventional RNNs for complex sequence modeling tasks.
  • Further research can optimize segment length for specific applications.