Neural computing architectures: The design of brain-like machines
Theoretical and applications aspects of neural-network (NN) computers are discussed in chapters contributed by European experts. Topics addressed include speech recognition based on topology-preserving neural maps, neural-map applications, backpropagation in nonfeedforward NNs, a parallel-distributed-processing learning approach to natural language, the learning capabilities of Boolean NNs, the logic of connectionist systems, and a probabilistic-logic NN for associative learning. Consideration is given to N-tuple sampling and genetic algorithms for speech recognition; the dynamic behavior of Boolean NNs; statistical mechanics and NNs; digital NNs, matched filters, and optical implementations; heteroassociative NNs using cabling vs link-disabling local modification rules; and the generation of movement trajectories in primates and robots. Also provided is an overview of parallel distributed processing.
- OSTI ID:
- 6445974
- Country of Publication:
- United States
- Language:
- English
Similar Records
Neural-network algorithms and architectures for pattern classification
Application of a convolutional neural network in permeability prediction: A case study in the Jacksonburg-Stringtown oil field, West Virginia, USA
Related Subjects
NEURAL NETWORKS
DESIGN
COMPUTER ARCHITECTURE
COMPUTER NETWORKS
COMPUTERS
CYBERNETICS
DIGITAL FILTERS
DISTRIBUTED DATA PROCESSING
MATHEMATICAL LOGIC
PARALLEL PROCESSING
PATTERN RECOGNITION
PROBABILITY
ROBOTS
SPEECH
STATISTICAL MECHANICS
USES
DATA PROCESSING
MECHANICS
PROCESSING
PROGRAMMING
990200* - Mathematics & Computers