Reservoir computing and the Sooner-is-Better bottleneck

Prior language input is not lost but integrated with the current input. This principle is demonstrated by "reservoir computing": Untrained recurrent neural networks project input sequences onto a random point in high-dimensional state space. Earlier inputs can be retrieved from this projec...

Full description

Saved in:
Bibliographic Details
Published in:The Behavioral and brain sciences 2016-01, Vol.39, p.e73-e73, Article e73
Main Authors: Frank, Stefan L, Fitz, Hartmut
Format: Article
Language:English
Subjects:
Aphasia
Auditory plasticity
Bismuth
Brain
Classification
Cognition & reasoning
Cognitive ability
Cognitive tasks
Computer applications
Computer simulation
Cortex
Energy conservation
Event-related potentials
Exposure
Firing pattern
Hippocampus
Humans
Information processing
Information systems
Inhibition
Language
Learning
Memory
Neostriatum
Nervous system
Neural networks
Neural Networks (Computer)
Neurons
Neurosciences
Nonlinear systems
Phonetics
Plasticity
Short term memory
Simulation
Synaptic plasticity
Task complexity
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Prior language input is not lost but integrated with the current input. This principle is demonstrated by "reservoir computing": Untrained recurrent neural networks project input sequences onto a random point in high-dimensional state space. Earlier inputs can be retrieved from this projection, albeit less reliably so as more input is received. The bottleneck is therefore not "Now-or-Never" but "Sooner-is-Better."
ISSN:0140-525X
1469-1825
DOI:10.1017/S0140525X15000783