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

Search research articles

Related Experiment Videos

Masking fields: a massively parallel neural architecture for learning, recognizing, and predicting multiple groupings

M A Cohen, S Grossberg

|May 11, 2010

Summary

This study introduces the masking field, a novel neural network architecture for pattern recognition. It efficiently processes complex inputs, offering solutions for credit assignment in cognitive systems.

Related Concept Videos

Parallel Processing

Parallel Processing

The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...

Masking and Demasking Agents

Masking and Demasking Agents

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 the metal...

You might also read

Related Articles

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

Sort by

Same author

De novo adamantinomatous craniopharyngioma with long-term pre-diagnostic imaging.

Neuro-Chirurgie·2023

Same author

The hippocampus and cerebellum in adaptively timed learning, recognition, and movement.

Journal of cognitive neuroscience·2013

Same author

Cortical synchronization and perceptual framing.

Journal of cognitive neuroscience·2013

Same author

A self-organizing neural model of motor equivalent reaching and tool use by a multijoint arm.

Journal of cognitive neuroscience·2013

Same author

Neural Representations for Sensorimotor Control. III. Learning a Body-Centered Representation of a Three-Dimensional Target Position.

Journal of cognitive neuroscience·2013

Same author

A case of pituitary apoplexy masquerading as hyperemesis gravidarum.

Journal of obstetrics and gynaecology : the journal of the Institute of Obstetrics and Gynaecology·2011

Area of Science:

* Computational Neuroscience
* Artificial Intelligence
* Cognitive Science

Background:

* Existing neural network architectures struggle with processing complex, multi-scale input patterns.
* The need for systems that can adapt sensitivity to pattern size while retaining microstructural detail is critical.
* Understanding the neural basis of learning and memory is essential for developing advanced cognitive models.

Purpose of the Study:

* To characterize a massively parallel neural network architecture, the masking field, through computer simulations.
* To demonstrate the masking field's ability to perform content-addressable memory functions and predictive coding.
* To present a stable design for the masking field and compare its mechanisms to neural substrates of learning and memory.

Main Methods:

Related Experiment Videos

* Systematic computer simulations of a cooperative-competitive feedback network (F(2)) receiving input from an adaptive filter (F(1)).
* Analysis of network behavior, including pattern recognition, code compression, and adaptive rescaling.
* Examination of the network's response to familiar versus unfamiliar input patterns and contextual information.

Main Results:

* The masking field successfully activates compressed, predictive recognition codes while masking unpredictitive ones.
* It demonstrates simultaneous detection of multiple groupings and adaptive rescaling of sensitivity to input pattern size.
* The network exhibits priming capabilities and adaptive sharpening, improving recognition accuracy with familiar patterns.

Conclusions:

* The masking field offers a solution to the credit assignment problem by real-time coding of predictive evidence.
* Its architecture, combining associative mechanisms, cooperative-competitive interactions, and gating signals, regulates learning of recognition codes.
* The masking field's capabilities are applicable to speech recognition, visual object recognition, and cognitive information processing.