Ãëàâíàÿ    Ex Libris    Êíèãè    Æóðíàëû    Ñòàòüè    Ñåðèè    Êàòàëîã    Wanted    Çàãðóçêà    ÕóäËèò    Ñïðàâêà    Ïîèñê ïî èíäåêñàì    Ïîèñê    Ôîðóì   
blank
Àâòîðèçàöèÿ

       
blank
Ïîèñê ïî óêàçàòåëÿì

blank
blank
blank
Êðàñîòà
blank
Mitchell T.M. — Machine Learning
Mitchell T.M. — Machine Learning



Îáñóäèòå êíèãó íà íàó÷íîì ôîðóìå



Íàøëè îïå÷àòêó?
Âûäåëèòå åå ìûøêîé è íàæìèòå Ctrl+Enter


Íàçâàíèå: Machine Learning

Àâòîð: Mitchell T.M.

Àííîòàöèÿ:

This book covers the field of machine learning, which is the study of algorithms that allow computer programs to automatically improve through experience. The book is intended to support upper level undergraduate and introductory level graduate courses in machine learning


ßçûê: en

Ðóáðèêà: Òåõíîëîãèÿ/

Ñòàòóñ ïðåäìåòíîãî óêàçàòåëÿ: Ãîòîâ óêàçàòåëü ñ íîìåðàìè ñòðàíèö

ed2k: ed2k stats

Ãîä èçäàíèÿ: 1997

Êîëè÷åñòâî ñòðàíèö: 432

Äîáàâëåíà â êàòàëîã: 21.06.2006

Îïåðàöèè: Ïîëîæèòü íà ïîëêó | Ñêîïèðîâàòü ññûëêó äëÿ ôîðóìà | Ñêîïèðîâàòü ID
blank
Ïðåäìåòíûé óêàçàòåëü
Find-S algorithm, mistake-bound learning in      220—221
Find-S algorithm, PAC learning of boolean conjunctions with      212
Find-S algorithm, search of hypothesis space      27—28
Finite horizon reward      371
First-order Horn clauses      283—284 318—319 See
First-order Horn clauses in analytical learning      311
First-order Horn clauses in Prolog-EBG      313 314
First-order logic, basic definitions      285
First-order representations, applications of      275
First-order resolution rule      296—297
First-order rules      274—275 283 301 302 See
First-order rules in FOIL algorithm      285—291
First-order rules, propositional rules, comparison with      283
Fitness function      250—252 255—256 258
Fitness proportionate selection      255
Fitness sharing      259
FOCL algorithm      302
FOCL algorithm, extensions to FOIL      357
FOCL algorithm, search step alteration with prior knowledge      339—340
FOIL algorithm      286 290—291 302
FOIL algorithm, extensions in FOCL      357
FOIL algorithm, information gain measure in      289
FOIL algorithm, Learn-one-rule and sequential covering algorithms, comparison with      287
FOIL algorithm, learning first-order rules in      285—291
FOIL algorithm, post-pruning in      291
FOIL algorithm, recursive rule learning in      290
Function approximation      8
Function approximation algorithms as lookup table substitute      384
Function approximation algorithms, choice of      9—11
Functions in logic      284 285
GABIL      256—259 269
GABIL, C4.5 and AQ14 algorithms, comparison with      256 258
GABIL, extensions to      258—259
GABIL, ID5R algorithm, comparison with      258
Gain ratio      73—74
Gas      See "Genetic algorithms"
Gaussian distribution      See "Normal distribution"
Gaussian kernel function      238—240
General-to-specific beam search      277—279 302
General-to-specific beam search in CN2 algorithm      278
General-to-specific beam search in FOCL algorithm      357—361
General-to-specific beam search in FOIL algorithm      287 357—358
General-to-specific beam search, advantages of      281
General-to-specific ordering of hypotheses      24—25 45—46 See
Generalization accuracy in neural networks      110—111
Generalizer      12 13
Generate-and-test beam search      250
Generate-and-test beam search, example-driven search, comparison with      281
Generate-and-test beam search, inverse entailment operators, comparison with      299
Generate-and-test beam search, inverse resolution, comparison with      298—299
Genetic algorithms      249—270
Genetic algorithms, advantages of      250
Genetic algorithms, applications of      256 269
Genetic algorithms, fitness function in      255—256
Genetic algorithms, limitations of      259
Genetic algorithms, parallelization of      268
Genetic algorithms, representation of hypotheses      252—253
Genetic algorithms, search of hypothesis space      259 268—269
Genetic operators      252—255 257 261—262
Genetic programming      250 262—266 269
Genetic programming, applications of      265 269
Genetic programming, performance of      266
Genetic programming, representation in      262—263
Gibbs algorithm      176
Global method      234
GOLEM      281
GP      See "Genetic programming"
Gradient ascent search      170—171
Gradient ascent search in Bayesian belief networks      188—190
Gradient descent search      89—91 93 97 115—116 123
Gradient descent search in EBNN algorithm      339
Gradient descent search, least-squared error hypothesis in      167
Gradient descent search, limitations of      92
Gradient descent search, weight update rule      91—92 237
Gradient descent search, weight update rule, stochastic approximation to      92—94 98—100 104—105 107—108
Gradient of error      91
Greedy search in Prolog-EBG      323
Greedy search in sequential covering algorithms      276—278
GRENDEL program      303
Ground literal      285
Halving algorithm      223
Halving algorithm, mistake-bound learning in      221—222
handwriting recognition      3—4
Handwriting recognition, Backpropagation algorithm in      81
Handwriting recognition, TangentProp algorithm in      348—349
Head of Horn clause      285
Hidden layer representations, discovery by Backpropagation algorithm      106—109 123
Hidden units in face recognition task      115—117
Hidden units, Backpropagation weight tuning rule for      103
Hidden units, Cascade-Correlation algorithm, addition by      121—123
Hidden units, choice in radial basis function networks      239—240
Hill-climbing search in FOIL algorithm      286 287
Hill-climbing search in genetic algorithms      268
Hill-climbing search in ID3 algorithm      60—61
Hoeffding bounds      210—211
Horn clauses      284 285
Horn clauses, first-order      See "First-order Horn clauses"
Human learning, explanations in      309
Human learning, prior knowledge in      330
Hypotheses      See also "Discrete-valued hypotheses" "General-to-specific "Hypothesis
Hypotheses, error differences between two      143—144
Hypotheses, estimation of accuracy      129—130
Hypotheses, estimation of accuracy, bias and variance in estimate      129 151 152
Hypotheses, estimation of accuracy, errors in      129—131 151
Hypotheses, evaluation of      128—129
Hypotheses, justification of, in inductive vs. analytical learning      334—336
Hypotheses, representations of      23
Hypotheses, testing of      144—145
Hypothesis space      14—15
Hypothesis space search by Backpropagation algorithm      97 106 122—123
Hypothesis space search by Backpropagation algorithm, comparison with decision tree learning      106
Hypothesis space search by Backpropagation algorithm, comparison with KBANN and TangentProp algorithms      350—351
Hypothesis space search by Find-S algorithm      27—28
Hypothesis space search by FOIL algorithm      286—287 357—361
Hypothesis space search by genetic algorithms      250 259
Hypothesis space search by gradient descent      90—91
Hypothesis space search by ID3 algorithm      60—62 64 76
Hypothesis space search by KBANN algorithm      346
Hypothesis space search by Learn-one-rule      277
Hypothesis space search by learning algorithms      24
Hypothesis space search in concept learning      23—25 46—47
Hypothesis space search in machine learning      14—15 18
Hypothesis space search, by Candidate-Elimination algorithm      64
Hypothesis space search, constraints on      302—303
Hypothesis space search, use of prior knowledge      339—340 362
Hypothesis space, bias in      40—42 46 129
Hypothesis space, finite, sample complexity for      207—214 225
Hypothesis space, infinite, sample complexity for      214—220
Hypothesis space, VC dimension of      214—217
ID3 algorithm      55—64 77
ID3 algorithm, backtracking and      62
ID3 algorithm, Candidate-Elimination algorithm, comparison with      61—62
ID3 algorithm, choice of attributes in      280—281
ID3 algorithm, choice of decision tree      63
ID3 algorithm, cost-sensitive measures      75—76
ID3 algorithm, extensions to      77 See
ID3 algorithm, inductive bias of      63—64 76
ID3 algorithm, Learn-one-rule, search comparison with      277
ID3 algorithm, limitations of      61—62
ID3 algorithm, overfitting in      67—68
ID3 algorithm, search of hypothesis space      60—62 64 76
ID3 algorithm, sequential covering algorithms, comparison with      280—281
ID3 algorithm, specialized for concept learning      56
ID3 algorithm, use of information gain in      58—60
ID5R algorithm, comparison with GABIL      258
ILP      See "Inductive logic programming"
Image encoding in face recognition      114
Imperfect domain theory in EBNN algorithm      356
Imperfect domain theory in explanation-based learning      330
Imperfect domain theory in FOCL algorithm      360
Imperfect domain theory in KBANN algorithm      344—345
Incremental explanation methods      328
Incremental gradient descent      See "Stochastic gradient descent"
Incremental Version Space Merging algorithm      47
Inductive bias      39—45 137—138 See "Preference "Restriction
Inductive bias in decision tree learning      63—66
Inductive bias in explanation-based learning      322—323
Inductive bias of Backpropagation algorithm      106
Inductive bias of Candidate-Elimination algorithm      43—46 63—64
Inductive bias of Find-S algorithm      45
Inductive bias of ID3 algorithm      63—64 76
Inductive bias of inductive learning algorithms      42—46
Inductive bias of k-Nearest Neighbor algorithm      234
Inductive bias of LMS algorithm      64
Inductive bias of Rote-Learner algorithm      44—45
Inductive bias, bias-free learning      40—42
Inductive bias, definition of      43
Inductive inference      See "Inductive learning"
Inductive learning      42 307—308 See "Genetic "Inductive "Neural
Inductive learning hypothesis      23
Inductive learning, analytical learning, comparison with      310 328—329 334—336 362
Inductive learning, inductive bias in      42—46
Inductive logic programming      275 291
Inductive logic programming, Prolog-EBG, comparison with      322
Inductive-analytical learning      334—363
Inductive-analytical learning, advantages of      362
Inductive-analytical learning, explanation-based learning and      330
Inductive-analytical learning, learning problem      337—338
Inductive-analytical learning, prior knowledge methods to alter search      339—340 362
Inductive-analytical learning, properties of ideal systems      337
Inductive-analytical learning, weighting of components in EBNN algorithm      351—352 355
Inductive-analytical learning, weighting prior knowledge in      338
Information gain      73
Information gain in FOIL algorithm      289
Information gain in ID3 algorithm      55 58—60
Information gain, definition of      57—58
Information theory, influence on machine learning      4
Information theory, Minimum Description Length principle and      172
Initialize-the-hypothesis approach      339—346
Initialize-the-hypothesis approach, Bayesian belief networks in      346
Instance-based learning      230—247 See "k-Nearest "Locally
Instance-based learning, advantages      245—246
Instance-based learning, case-based reasoning, comparison with other methods      240
Instance-based learning, limitations of      231
Inverse entailment      292 302
Inverse entailment in Progol      300—302
Inverse entailment, first-order      297
Inverse entailment, generate-and-test beam search, comparison with      299
Inverse resolution      294—296 302
Inverse resolution, first-order      297—298
Inverse resolution, generate-and-test beam search, comparison with      298—299
Inverse resolution, limitations of      300
Inverted deduction      291—293
Jacobian      354
Job-shop scheduling, genetic algorithms in      256
Joint probability distribution in Bayesian belief networks      185—187
k-fold cross-validation      112 147 150
k-means problem      191—193
k-means problem, derivation of EM algorithm for      195—196
k-nearest neighbor algorithm      231—233 246
k-Nearest Neighbor algorithm, applications of      234
k-Nearest Neighbor algorithm, cross-validation in      235
k-Nearest Neighbor algorithm, decision tree and rule learning, comparison with      235
k-Nearest Neighbor algorithm, distance-weighted      233—234
k-Nearest Neighbor algorithm, inductive bias of      234
k-Nearest Neighbor algorithm, memory indexing in      236
k-term CNF expressions      213—214
k-term DNF expressions      213—214
K2 algorithm      190—191
KBANN algorithm      340—347 362 387
KBANN algorithm, advantages of      344
KBANN algorithm, Backpropagation algorithm, comparison with      344—345
KBANN algorithm, Backpropagation weight update rule in      343—344
KBANN algorithm, hypothesis space search by Backpropagation and TangentProp, comparison with      350—351
KBANN algorithm, limitations of      345
KBANN algorithm, prior knowledge in      339
kd-tree      236
Kernel function      236 238 246
Kernel function, Gaussian      See "Gaussian kernel function"
Knowledge compilation      320
Knowledge level learning      323—325
Knowledge reformulation      320
Knowledge-Based Artificial Neural Network (KBANN) algorithm      See "KBANN algorithm"
Lamarckian evolution      266
Language bias      See "Restriction bias"
Lazy explanation methods      328
Lazy learning methods, comparison with eager learning      244—245
Learn-one-rule algorithm, FOIL algorithm, comparison with      287
Learn-one-rule algorithm, ID3 algorithm, search comparison with      277
Learn-one-rule algorithm, rule performance in      282
Learn-one-rule algorithm, rule post-pruning in      281
Learn-one-rule algorithm, variations of      279—280 286
Learning algorithms, consistent learners      162—163
Learning algorithms, design of      9—11 17
Learning algorithms, domain-independent      336
Learning algorithms, error differences between two      145—151
Learning algorithms, search of hypothesis space      24
Learning problems      2—5 17
Learning problems in inductive-analytical learning      337—338
Learning problems, computational theory of      201—202
Learning rate      88 91
Learning systems, design of      5—14 17
Learning systems, program modules      11—14
Learning, human      See "Human learning"
Learning, machine      See "Machine learning"
Learn_naive_Bayes_text      182—183
Least mean squares algorithm      See "LMS algorithm"
Least-squared error hypothesis, classifiers for      198
Least-squared error hypothesis, gradient descent in      167
Least-squared error hypothesis, maximum likelihood (ML) hypothesis and      164—167
Leave-one-out cross-validation      235
Legal case reasoning, case-based reasoning in      240
Lemma-Enumerator algorithm      324
Lifelong learning      370
Line search      119
Linear programming, as weight update algorithm      95
Linearly separable sets      86 89 95
List-Then-Eliminate algorithm      30
Literal      284 285
LMS algorithm      11 15
LMS algorithm, inductive bias of      64
LMS weight update rule      See "Delta rule"
Local method      234
Locally weighted regression      231 236—238 246
Locally weighted regression, limitations of      238
Locally weighted regression, weight update rules in      237—238
logical constants      284 285
Logical terms      284 285
Logistic function      96 104
Lookup table, function approximation algorithms as substitute      384
Lookup table, neural network as substitute      384
Lower bound on sample complexity      217—218
m-estimate of probability      179—180 198 282
Machine learning      15 See
Machine learning, applications      5 17
Machine learning, definition of      2
Machine learning, influence of other disciplines on      4 17
Machine learning, search of hypothesis space      14—15 18
Manufacturing process control      17
MAP hypothesis      See "Maximum a posteriori hypothesis"
MAP Learning algorithm, Brute-Force      See "Brute-Force MAP Learning algorithm"
Markov decision processes (MDP)      370 387
Markov decision processes (MDP), applications of      386
Markus      302
Marvin      302
Maximally general hypotheses, computation by Candidate-Elimination algorithm      31 46
Maximally specific hypotheses, computation by Candidate-Elimination algorithm      31 46
Maximally specific hypotheses, computation by Find-S algorithm      26—28 62—63
Maximum a posteriori (MAP) hypothesis      157 197 See
Maximum a posteriori (MAP) hypothesis, naive Bayes classifier and      178
1 2 3 4
blank
Ðåêëàìà
blank
blank
HR
@Mail.ru
       © Ýëåêòðîííàÿ áèáëèîòåêà ïîïå÷èòåëüñêîãî ñîâåòà ìåõìàòà ÌÃÓ, 2004-2024
Ýëåêòðîííàÿ áèáëèîòåêà ìåõìàòà ÌÃÓ | Valid HTML 4.01! | Valid CSS! Î ïðîåêòå