Full width home advertisement

Travel the world

Climb the mountains

Post Page Advertisement [Top]

Artificial IntelligencebookMachine Learning

Understanding Machine Learning: From Theory to Algorithms By Shalev-Shwartz S. & Ben-David S.

Understanding Machine Learning: From Theory to Algorithms By Shalev-Shwartz S. & Ben-David S. PDF

spinningbot-temp-inter-1

Informations about the book:

TitleUnderstanding Machine Learning: From Theory to Algorithms

AuthorShalev-Shwartz S. & Ben-David S.

Size: 3 Mb

Format: PDF

Year: 2014

Pages: 416

Book Contents:

1 Introduction  
1.1 What Is Learning?  
1.2 When Do We Need Machine Learning? 
1.3 Types of Learning  
1.4 Relations to Other Fields  
1.5 How to Read This Book 
1.6 Notation 
Part 1 Foundations  
2 A Gentle Start  
2.1 A Formal Model – The Statistical Learning Framework  
2.2 Empirical Risk Minimization  
2.3 Empirical Risk Minimization with Inductive Bias  
2.4 Exercises  
3 A Formal Learning Model  
3.1 PAC Learning  
3.2 A More General Learning Model 
3.4 Bibliographic Remarks  
3.5 Exercises  
4 Learning via Uniform Convergence  
4.1 Uniform Convergence Is Sufficient for Learnability  
4.2 Finite Classes Are Agnostic PAC Learnable  
4.4 Bibliographic Remarks  
4.5 Exercises 
5 The Bias-Complexity Tradeoff  
5.1 The No-Free-Lunch Theorem  
5.2 Error Decomposition   
5.4 Bibliographic Remarks  
5.5 Exercises  
6 The VC-Dimension  
6.1 Infinite-Size Classes Can Be Learnable  
6.2 The VC-Dimension  
6.3 Examples  
6.4 The Fundamental Theorem of PAC learning  
6.5 Proof of Theorem 6.7  
6.7 Bibliographic remarks  
6.8 Exercises  
7 Nonuniform Learnability  
7.1 Nonuniform Learnability  
7.2 Structural Risk Minimization  
7.3 Minimum Description Length and Occam’s Razor  
7.4 Other Notions of Learnability – Consistency  
7.5 Discussing the Different Notions of Learnability  
7.7 Bibliographic Remarks  
7.8 Exercises  
8 The Runtime of Learning  
8.1 Computational Complexity of Learning  
8.2 Implementing the ERM Rule  
8.3 Efficiently Learnable, but Not by a Proper ERM  
8.4 Hardness of Learning  
8.6 Bibliographic Remarks  
8.7 Exercises  
Part 2 From Theory to Algorithms  
9 Linear Predictors  
9.1 Halfspaces  
9.2 Linear Regression  
9.3 Logistic Regression  
9.5 Bibliographic Remarks  
9.6 Exercises   
10 Boosting  
10.1 Weak Learnability  
10.2 AdaBoost  
10.3 Linear Combinations of Base Hypotheses  
10.4 AdaBoost for Face Recognition  
10.6 Bibliographic Remarks  
10.7 Exercises  
11 Model Selection and Validation  
11.1 Model Selection Using SRM  
11.2 Validation  
11.3 What to Do If Learning Fails  
11.5 Exercises  
12 Convex Learning Problems  
12.1 Convexity, Lipschitzness, and Smoothness  
12.2 Convex Learning Problems  
12.3 Surrogate Loss Functions  
12.5 Bibliographic Remarks  
12.6 Exercises  
13 Regularization and Stability  
13.1 Regularized Loss Minimization  
13.2 Stable Rules Do Not Overfit  
13.3 Tikhonov Regularization as a Stabilizer  
13.4 Controlling the Fitting-Stability Tradeoff 
13.6 Bibliographic Remarks  
13.7 Exercises  
14 Stochastic Gradient Descent  
14.1 Gradient Descent  
14.2 Subgradients  
14.3 Stochastic Gradient Descent (SGD)  
14.4 Variants  
14.5 Learning with SGD 
14.7 Bibliographic Remarks  
14.8 Exercises  
15 Support Vector Machines  
15.1 Margin and Hard-SVM  
15.2 Soft-SVM and Norm Regularization  
15.3 Optimality Conditions and “Support Vectors”  
15.4 Duality   
15.5 Implementing Soft-SVM Using SGD  
15.7 Bibliographic Remarks  
15.8 Exercises  
16 Kernel Methods  
16.1 Embeddings into Feature Spaces  
16.2 The Kernel Trick  
16.3 Implementing Soft-SVM with Kernels  
16.5 Bibliographic Remarks  
16.6 Exercises  
17 Multiclass, Ranking, and Complex Prediction Problems 190
17.1 One-versus-All and All-Pairs 190
17.2 Linear Multiclass Predictors 193
17.3 Structured Output Prediction 198
17.4 Ranking  
17.5 Bipartite Ranking and Multivariate Performance Measures  
17.7 Bibliographic Remarks  
17.8 Exercises  
18 Decision Trees  
18.1 Sample Complexity  
18.2 Decision Tree Algorithms  
18.3 Random Forests   
18.5 Bibliographic Remarks  
18.6 Exercises  
19 Nearest Neighbor  
19.1 k Nearest Neighbors  
19.2 Analysis  
19.3 Efficient Implementation  
19.5 Bibliographic Remarks  
19.6 Exercises  
20 Neural Networks  
20.1 Feedforward Neural Networks  
20.2 Learning Neural Networks  
20.3 The Expressive Power of Neural Networks  
20.4 The Sample Complexity of Neural Networks  
20.5 The Runtime of Learning Neural Networks  
20.6 SGD and Backpropagation 
20.8 Bibliographic Remarks  
20.9 Exercises  
Part 3 Additional Learning Models  
21 Online Learning  
21.1 Online Classification in the Realizable Case  
21.2 Online Classification in the Unrealizable Case  
21.3 Online Convex Optimization  
21.4 The Online Perceptron Algorithm  
21.6 Bibliographic Remarks  
21.7 Exercises  
22 Clustering  
22.1 Linkage-Based Clustering Algorithms  
22.2 k-Means and Other Cost Minimization Clusterings  
22.3 Spectral Clustering  
22.4 Information Bottleneck  
22.5 A High Level View of Clustering 
22.7 Bibliographic Remarks  
22.8 Exercises  
23 Dimensionality Reduction  
23.1 Principal Component Analysis (PCA)  
23.2 Random Projections  
23.3 Compressed Sensing  
23.4 PCA or Compressed Sensing? 
23.6 Bibliographic Remarks  
23.7 Exercises  
24 Generative Models  
24.1 Maximum Likelihood Estimator  
24.2 Naive Bayes  
24.3 Linear Discriminant Analysis  
24.4 Latent Variables and the EM Algorithm  
24.5 Bayesian Reasoning   
24.7 Bibliographic Remarks  
24.8 Exercises  
25 Feature Selection and Generation  
25.1 Feature Selection  
25.2 Feature Manipulation and Normalization  
25.3 Feature Learning  
25.5 Bibliographic Remarks  
25.6 Exercises  
Part 4 Advanced Theory  
26 Rademacher Complexities  
26.1 The Rademacher Complexity  
26.2 Rademacher Complexity of Linear Classes  
26.3 Generalization Bounds for SVM  
26.4 Generalization Bounds for Predictors with Low   Norm  
26.5 Bibliographic Remarks  
27 Covering Numbers  
27.1 Covering  
27.2 From Covering to Rademacher Complexity via Chaining  
27.3 Bibliographic Remarks  
28 Proof of the Fundamental Theorem of Learning Theory  
28.1 The Upper Bound for the Agnostic Case  
28.2 The Lower Bound for the Agnostic Case  
28.3 The Upper Bound for the Realizable Case  
29 Multiclass Learnability  
29.1 The Natarajan Dimension  
29.2 The Multiclass Fundamental Theorem  
29.3 Calculating the Natarajan Dimension  
29.4 On Good and Bad ERMs  
29.5 Bibliographic Remarks  
29.6 Exercises  
30 Compression Bounds  
30.1 Compression Bounds  
30.2 Examples  
30.3 Bibliographic Remarks  
31 PAC-Bayes  
31.1 PAC-Bayes Bounds  
31.2 Bibliographic Remarks  
31.3 Exercises  
Appendix A Technical Lemmas  
Appendix B Measure Concentration  
B.1 Markov’s Inequality  
B.2 Chebyshev’s Inequality  
B.3 Chernoff’s Bounds  
B.4 Hoeffding’s Inequality  
B.5 Bennet’s and Bernstein’s Inequalities  
B.6 Slud’s Inequality  
B.7 Concentration of χ2 Variables  
Appendix C Linear Algebra  
C.1 Basic Definitions  
C.2 Eigenvalues and Eigenvectors  
C.3 Positive definite matrices  
C.4 Singular Value Decomposition (SVD) 

png+button

No comments:

Post a Comment

Bottom Ad [Post Page]