Выбрать главу

statistical, 37, 228, 297, 300, 307

statistical relational, 309

supervised, 57-202, 209, 214, 220, 222, 226

unsupervised, 203-233

Learning algorithms (learners)

Alchemy, 250-259

children’s learning and, 203-204

control of, 45

curse of dimensionality and, 188

empirical evaluation of, 76

evaluation and, 283

interaction with, 264-267

machine learning and, 6-10

optimization and, 283

prediction and, 39

representation and, 283

speed and, 139-142

as superpredators, 8-9

variety of tasks undertaken by, 23-25

See also individual algorithms

LeCun, Yann, 113, 195

Lehman Brothers, 106

Leibniz, Gottfried, 58, 64, 175, 198

Lenat, Doug, 35

Lewis, Michael, 39

Linear regression, 15, 50, 113, 182, 214, 306

LinkedIn, 269, 271

Lipson, Hod, 121

Local minimum, 102, 103, 110-111

Local optima, 111, 128

Locally weighted regression, 182, 306

Locke, John, 58, 93, 178

Logic, 50, 33, 49, 80-81

Bayesians and, 173

computers and, 2

Master Algorithm and, 240, 244, 245-246

probability and, 173-175, 245-246, 306, 309

unified with graphical models, 245-250

Logical inference, Alchemy and, 256

Logic gates, 96

Logistic curve. See S curves

Long-tail phenomenon, 12, 299

Long-term potentiation, 27

Loopy belief propagation, 163-164, 231

Lorenz, Konrad, 138

Low-pass filter, 133

Machine learners, knowledge engineers vs., 34-38

Machine learning, 6-10

analogy and, 178-179

bias and variance and, 78-79

big data and, 15-16

business and, 10-13

chunking, 223-227

clustering, 205-210

dimensionality reduction, 211-217

effect on employment, 276-279

exponential function and, 73-74

fitness function and, 123

further readings, 297-298

future of, 21-22

impact on daily life, 298

effect on employment, 276-279

meta-learning, 237-239

nature vs. nurture debate and, 29, 137-139

Newton’s principle and, 65-66

planetary-scale, 256-259

politics and, 16-19

principal-component analysis, 211-217

problem of unpredictability and, 38-40

reinforcement learning, 218-223, 226-227

relational learning, 227-233

relationship to artificial intelligence, 8

science and, 13-16, 235-236

significance tests and, 76-77

as technology, 236-237

Turing point and, 286, 288

war and, 19-21, 279-282

See also Algorithms

Machine-learning problem, 61-62, 109-110

Machine-translation systems, 154

MacKay, David, 170

Madrigal, Alexis, 273-274

Malthus, Thomas, 178, 235

Manchester Institute of Biotechnology, 16

Mandelbrot set, 30, 300

Margins, 192-194, 196, 241, 242, 243, 307

Markov, Andrei, 153

Markov chain Monte Carlo (MCMC), 164-165, 167, 170, 231, 241, 242, 253, 256

Markov chains, 153-155, 159, 304-305

Markov logic. See Markov logic networks (MLNs)

Markov logic networks (MLNs), 246-259, 309-310

classes and, 257

complexity and, 258-259

parts and, 256-257

with hierarchical structure, 256-257

See also Alchemy

Markov networks, 171-172, 229, 240, 245, 253, 306

Marr, David, 89

Marr’s three levels, 89

Master Algorithm, 239-246

Alchemy and, 250-259

Bayes’ theorem and, 148

brain as, 26-28

CanceRx, 259-261

candidates that fail as, 48-50

chunking and, 226

complexity of, 40-41

as composite picture of current and future learners, 263-264

computer science and, 32-34

equation, 50

evolution and, 28-29

five tribes and, 51-55

future and, 292

goal of, 39

Google and, 282

invention of, 25-26

Markov logic networks and, 236-250

meta-learning and, 237-239

physics and, 29-31

practical applications of, 41-45

statistics and, 31-32

symbolism and, 90-91

theory of everything and, 46-48

Turing point and, 286, 288

as unifier of machine learning, 237

unity of knowledge and, 31

Match.com, 12, 265

Matrix factorization for recommendation systems, 215

Maximum likelihood principle, 166-167, 168

Maxwell, James Clerk, 235

McCulloch, Warren, 96

McKinsey Global Institute, 9

MCMC. See Markov chain Monte Carlo (MCMC)

Means-ends analysis, 225

Mechanical Turk, 14

Medical data, sharing of, 272-273

Medical diagnosis, 23, 147, 149-150, 160, 169, 228-229, 248-249

Memorization, 48

Memory, time as principal component of, 217

Mencken, H. L., 230

Mendeleev, Dmitri, 235

Meta-learning, 237-239, 255, 309

Methane/methanol, 197-198

Michalski, Ryszard, 69, 70, 90

Michelangelo, 2

Microprocessor, 48-49, 236

Microsoft, 9, 22

Kinect, 88, 237, 238

Windows, 12, 133, 224

Xbox Live, 160-161

Microsoft Research, 152

Military robots, 21, 279-282, 299, 310

Mill, John Stuart, 93

Miller, George, 224

Minsky, Marvin, 35, 38, 100-101, 102, 110, 112, 113

Mitchell, Tom, 64, 69, 90

Mixability, 135

MLNs. See Markov logic networks (MLNs)

Moby Dick (Melville), 72

Molecular biology, data and, 14

Moneyball (Lewis), 39

Mooney, Ray, 76

Moore’s law, 287

Moravec, Hans, 288

Muggleton, Steve, 80

Multilayer perceptron, 108-111

autoencoder, 116-118

Bayesian, 170

driving a car and, 113

Master Algorithm and, 244

NETtalk system, 112

reinforcement learning and, 222

support vector machines and, 195

Music composition, case-based reasoning and, 199

Music Genome Project, 171

Mutation, 124, 134-135, 241, 252

Naïve Bayes classifier, 151-153, 171, 304

Bayesian networks and, 158-159

clustering and, 209

Master Algorithm and, 245

medical diagnosis and, 23

relational learning and, 228-229

spam filters and, 23-24

text classification and, 195-196

Narrative Science, 276

National Security Agency (NSA), 19-20, 232

Natural selection, 28-29, 30, 52

as algorithm, 123-128

Nature

Bayesians and, 141

evolutionaries and, 137-142

symbolists and, 141

Nature (journal), 26

Nature vs. nurture debate, machine learning and, 29, 137-139

Neal, Radford, 170

Nearest-neighbor algorithms, 24, 178-186, 202, 306-307

dimensionality and, 186-190

Negative examples, 67

Netflix, 12-13, 183-184, 237, 266

Netflix Prize, 238, 292

Netscape, 9

NETtalk system, 112

Network effect, 12, 299

Neumann, John von, 72, 123

Neural learning, fitness and, 138-139

Neural networks, 99, 100, 112-114, 122, 204

convolutional, 117-118, 302-303

Master Algorithm and, 240, 244, 245

reinforcement learning and, 222

spin glasses and, 102-103

Neural network structure, Baldwin effect and, 139

Neurons

action potentials and, 95-96, 104-105

Hebb’s rule and, 93-94

McCulloch-Pitts model of, 96-97

processing in brain and, 94-95

See also Perceptron

Neuroscience, Master Algorithm and, 26-28

Newell, Allen, 224-226, 302

Newhouse, Neil, 17

Newman, Mark, 160

Newton, Isaac, 293

attribute selection, 189

laws of, 4, 14, 15, 46, 235

rules of induction, 65-66, 81, 82