Page 327 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

320 Index

Meaning vs. information, 282e284 MRI. See Magnetic resonance imaging (MRI)
Media guide, 107 MTM. See Medium-term memory (MTM)
Medium-term memory (MTM), 45 Multichannel neurophysiological
MEG. See Magnetoencephalography (MEG) measurements, 226
Memory, prediction vs., 284e287 Multilayer neural network, 18
Mental models, 87 Multilayer perceptron (MLP), 220e221
Mesoscopic coupling, 213e214 Multimodal brain parcellation, 275
“Meta data”, 144 Multimodal neuroimaging feature learning with
Metaanalysis, 266e267, 269 DL, 276e277
Metacognitive neurofuzzy inference system Multiple faults, 256
(McFIS), 120 Multistability, 210
Metaphorical Brain 2, 81 in physics and biology, 211e215, 213f
Metastable/metastability Multivariate Gaussian random dataset, 21
behaviors, 211e213 Multiview genomic data integration methodology
in cognition and brain dynamics, 210e211, 212f (MVDA methodology), 270e272, 271f
cognitive states, 211 Multiview learning, 265e269. See also Deep
patterns, 211, 215 learning (DL)
MFE. See Minimum free energy (MFE) analysis types, 269
Mild Cognitive Impairment (MCI), 236 in bioinformatics, 269e273
Mind, computational theory of, 83e84. See also data integration taxonomy, 267f
Brain-mind-computer trichotomy data type, 268
Minimal anatomies method, 33e34 deep multimodal feature learning, 275e277
Minimum free energy (MFE), 55e56, 60, 63e64 integration stage, 267e268, 268f
gradient descent, 70e74 multiview data related to clinical tests, 266f
unsupervised learning rule, 70e74 in neuroinformatics, 273e275
Minimum Helmholtz free energy. See Minimum MVDA methodology. See Multiview genomic
free energy (MFE) data integration methodology (MVDA
Mirror neurons, 87 methodology)
Misaligned objects, 285e286
MLCI. See Mouse-level computational N
intelligence (MLCI) NASA, 57
MLP. See Multilayer perceptron (MLP) National Science Foundation of China (NSFC), 162
Model-free fault diagnosis systems, 254e256 Natural evolution, 116
research challenges, 256 Natural intelligence (NI), 56, 60e61
Modeling aspect, 256 Nature’s learning rule
Modular connectionist-based systems, 117 Adaline
Module chromosome, 297e298 and LMS algorithm, 3e5
Monitoring system, 257 unsupervised learning with, 5e6
Motor control, 194 bootstrap learning
Motor theory of speech, 101 with more “biologically correct” sigmoidal
Mouse brain to human mind, 182e184 neuron, 13e20
Mouse-level computational intelligence with sigmoidal neuron, 10e12
(MLCI), 179 clustering algorithms, 20e21
from RNNs to Hebbian-LMS algorithm
deep vs. broad, 175e176 general, 21e22
emerging hardware to enhance capability by nature’s, 26e27
orders of magnitude, 176e178 postulates and, 26
recurrent neural network, 172e175 trainable neural network incorporating,
roadmap for, 176 27e29
MPD computing architecture. See Massively postulates of synaptic plasticity, 25
parallel distributed computing architecture Robert Lucky’s adaptive equalization, 7e10
(MPD computing architecture) synapse, 22e25