forked from neonwatty/machine-learning-refined
-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathindex.html
351 lines (262 loc) · 34.4 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
<!DOCTYPE html>
<html>
<head>
<!-- always use https, not http -->
<script type="text/javascript" src='https://ajax.googleapis.com/ajax/libs/jquery/1.10.1/jquery.min.js'></script>
<link rel="stylesheet" href="html/CSS/home.css">
<!-- need this for navigation menu -->
<link rel="stylesheet" href="html/CSS/navbar.css">
<!-- subscription -->
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.4.0/css/font-awesome.min.css">
<link rel="stylesheet" href="html/CSS/subscription.css">
</head>
<body>
<!-- Navigation menu -->
<div class ="navmenu">
<ul>
<li class="menu-item active"><a href="index.html"><span class="menu-item">HOME</span></a></li>
<!-- <li class="menu-item"><a href="html/pages/presentations.html"><span class="menu-item">PRESENTATIONS</span></a></li>
<li class="menu-item"><a href="html/pages/gallery.html"><span class="menu-item">GALLERY</span></a></li> -->
<li class="menu-item"><a href="html/pages/random_thoughts.html"><span class="menu-item">RANDOM THOUGHTS</span></a></li>
<li class="menu-item"><a href="html/pages/about.html"><span class="menu-item">ABOUT</span></a></li>
</ul>
</div>
<!-- still image background
<div class="bg">
</div>
-->
<video autoplay muted loop id="myVideo" src="html/vids/background_v1.mov"></video>
<script>
var vid = document.getElementById("myVideo");
vid.playbackRate = 1.5;
</script>
<!-- <p id="title">Machine Learning Refined</p> -->
<br/><br/>
<p class="sub-title"> This is a blog about machine learning and deep learning fundamentals built by the authors of the
textbook <a class="redlink" target="_blank" href="http://mlrefined.com">Machine Learning Refined</a>
published by Cambridge University Press. The posts, cut into short series, use careful writing and interactive coding widgets
to provide an intuitive and playful way to learn about core concepts in AI - from some of the most basic to the most advanced.
<br/><br/>
Each and every post here is a Python Jupyter notebook, prettied up for the web, that you can download and run on your own machine
by pulling our <a class="redlink" target="_blank" href="https://github.com/jermwatt/mlrefined">GitHub repo</a>.
<br/><br/>
© The material on this website is not to be distributed, copied, or reused without written permission from the authors.
</p>
<br/>
<hr style="display:block; margin-top: 0.5em; margin-bottom: 0.5em; margin-left: 240px; margin-right: 240px; border-style: inset; border-width: 1px;"/>
<br/>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter03" />
<label for="chapter03"><span>CHAPTER 3. Derivatives and automatic differentiation</span></label>
<div class="twelve_col">
<div class="description">
<p><strong>3.1. </strong> What are derivatives? <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_1_what_are_derivatives.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.2. </strong> Derivatives at a point and the numerical differentiator <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_2_numerical_differentiator.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>3.3. </strong> Derivative equations and hand computations <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_3_derivative_equations.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>3.4. </strong> Automatic differentiation - the forward mode <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_4_AD_forward_mode.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.5. </strong> Higher order derivatives <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_5_higher_order.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.6. </strong> Taylor series <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_6_Taylor_Series.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.7. </strong> Derivatives of multi-input functions <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_7_multi_input_derivatives.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.8. </strong> Automatic differentiation for multi-input functions <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_8_AD_multi_input_functions.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.9. </strong> The reverse mode of automatic differentiation <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_9_AD_reverse_mode.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.10. </strong> The Hessian and higher order derivatives <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_10_higher_multivariate.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.11. </strong> Multi-input Taylor Series <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_11_multi_taylor_series.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
<p><strong>3.12. </strong> Getting to know autograd: your professional grade automatic differentiator <a target="_blank" class="sublink-active" href="blog_posts/3_Automatic_differentiation/3_12_getting_to_know_autograd.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a> </p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter05" />
<label for="chapter05"><span>CHAPTER 5. Optimization: zero order methods </span></label>
<div class="six_col">
<div class="description">
<p><strong>5.0. </strong> Motivation for mathematical optimization<a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_0_Motivation.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/Mathematical_Optimization_presentations/Part_1_motivation_random_search.slides.html#/"> slides</a></p>
<p><strong>5.1. </strong> The zero order condition for optimality <a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_1_zero_order_condition.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>5.2. </strong> Global optimization methods <a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_2_Global_optimization_methods.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>5.3. </strong> Local optimization methods <a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_3_Local_optimization_methods.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>5.4. </strong> Random search<a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_4_Random_search.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>5.5. </strong> Coordinate search and descent<a target="_blank" class="sublink-active" href="blog_posts/5_Zero_order_methods/5_5_Coordinate_search_and_descent.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter06" checked/>
<label for="chapter06"><span>CHAPTER 6. Optimization: first order methods</span></label>
<div class="seven_col">
<div class="description">
<p><strong>6.0. </strong> Introduction <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_0_Introduction.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>6.1. </strong> The first order optimality condition <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_1_First_order_condition.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>6.2. </strong> The geometric anatomy of lines and hyperplanes <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_2_Hyperplane_anatomy.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/6_First_order_methods/6_2_Hyperplane_anatomy.slides.html#/"> slides</a></p>
<p><strong>6.3. </strong> The geometric anatomy of first order Taylor series approximations <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_3_tangent_plane_anatomy.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/6_First_order_methods/6_3_tangent_plane_anatomy.slides.html#/"> slides</a></p>
<p><strong>6.4. </strong> Gradient descent <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_4_Gradient_descent.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/6_First_order_methods/6_4_Gradient_descent.slides.html#/"> slides</a></p>
<p><strong>6.5. </strong> Conservative steplength rules <a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_5_Conservative_steplength_rules.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>6.6. </strong> First order coordinate descent methods<a target="_blank" class="sublink-active" href="blog_posts/6_First_order_methods/6_6_Coordinate_descent.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter07" checked/>
<label for="chapter07"><span>CHAPTER 7. Optimization: second order methods</span></label>
<div class="eight_col">
<div class="description">
<p><strong>7.1. </strong> Quadratic functions <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_1_Quadratic_functions.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.2. </strong> Second order derivatives and curvature <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_2_Second_derivatives_curvature.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.3. </strong> Newton's method <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_3_Newtons_method.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.4. </strong> Regularization, Newton's method, and non-convex functions<a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_4_Newtons_regularized.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.5. </strong> The first order derivation of Newton's method <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_5_Newtons_firstorder.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.6. </strong> Quasi-Newton methods <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_6_Quasi_newton_methods.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.7. </strong> Coordinate descent <a target="_blank" class="sublink-inactive" href=""> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>7.8. </strong> The second order optimality condition <a target="_blank" class="sublink-active" href="blog_posts/7_Second_order_methods/7_8_Second_order_condition.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter08" checked/>
<label for="chapter08"><span>CHAPTER 8. Linear regression </span></label>
<div class="four_col">
<div class="description">
<p><strong>8.1. </strong> Least squares linear regression <a target="_blank" class="sublink-active" href="blog_posts/8_Linear_regression/8_1_Least_squares_regression.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>8.2. </strong> The probabilistic perspective on Least Squares linear regression <a target="_blank" class="sublink-active" href="blog_posts/8_Linear_regression/8_2_Probabilistic_least_squares.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>8.3. </strong> Least absolute deviations linear regression <a target="_blank" class="sublink-active" href="blog_posts/8_Linear_regression/8_3_Least_absolute_deviations.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>8.4. </strong> Feature scaling via standard normalization <a target="_blank" class="sublink-active" href="blog_posts/8_Linear_regression/8_4_feature_scaling.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/8_Linear_regression/8_4_feature_scaling_SLIDES.slides.html"> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter09" checked/>
<label for="chapter09"><span>CHAPTER 9. Linear two class classification </span></label>
<div class="four_col">
<div class="description">
<p><strong>9.1. </strong> Logistic regression <a target="_blank" class="sublink-active" href="blog_posts/9_Linear_twoclass_classification/9_1_Logistic_regression.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/courses/deep_learning/Lecture_3_logistic_regression.slides.html#/"> slides</a></p>
<p><strong>9.2. </strong> The perceptron <a target="_blank" class="sublink-active" href="blog_posts/9_Linear_twoclass_classification/9_2_Perceptron.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>9.3. </strong> Support vector machines <a target="_blank" class="sublink-inactive" href=""> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>9.4. </strong> Feature scaling via standard normalization <a target="_blank" class="sublink-active" href="blog_posts/9_Linear_twoclass_classification/9_4_Feature_scaling.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter10" checked/>
<label for="chapter10"><span>CHAPTER 10. Linear multiclass classification </span></label>
<div class="five_col">
<div class="description">
<p><strong>10.1. </strong> One-versus-All classification <a target="_blank" class="sublink-active" href="blog_posts/10_Linear_multiclass_classification/10_1_One_versus_all.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/courses/deep_learning/Lecture_5_one_versus_all.slides.html#/"> slides</a></p>
<p><strong>10.2. </strong> Multiclass softmax classification <a target="_blank" class="sublink-active" href="blog_posts/10_Linear_multiclass_classification/10_2_Multiclass_classification.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/courses/deep_learning/Lecture_6_multiclass_classification.slides.html#/"> slides</a></p>
<p><strong>10.3. </strong> Multiclass classification with categorical labels <a target="_blank" class="sublink-inactive" href="blog_posts/10_Linear_multiclass_classification/10_3_Feature_scaling.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>10.4. </strong> Feature scaling via standard normalization <a target="_blank" class="sublink-active" href="blog_posts/10_Linear_multiclass_classification/10_4_Feature_scaling.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>10.5. </strong> Stochastic and mini-batch gradient descent <a target="_blank" class="sublink-inactive" href="blog_posts/10_Linear_multiclass_classification/10_4_Feature_scaling.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter11" checked/>
<label for="chapter11"><span>CHAPTER 11. Linear unsupervised learning </span></label>
<div class="six_col">
<div class="description">
<p><strong>11.1. </strong> Fixed spanning sets, orthonormality, and projections <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_1_Spanning_sets_orthonormality_projections.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>11.2. </strong> Principal Component Analysis and the Autoencoder <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_2_Principal_Component_Analysis.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>11.3. </strong> Feature scaling via PCA sphereing <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_3_PCA_sphereing.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>11.4. </strong> Recommender Systems <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_4_Recommender_Systems.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>11.5. </strong> K-means clustering <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_5_K_means_clustering.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>11.6. </strong> General matrix factorization techniques <a target="_blank" class="sublink-active" href="blog_posts/11_Linear_unsupervised_learning/11_6_matrix_factorization.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter12" checked/>
<label for="chapter12"><span>CHAPTER 12. Introduction to nonlinear learning</span></label>
<div class="ten_col">
<div class="description">
<p><strong>12.1. </strong> Features, functions, and nonlinear regression <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_1_Introduction_nonlinear_regression.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/12_Nonlinear_intro/12_1_Introduction_nonlinear_regression_SLIDES.slides.html#/"> slides</a></p>
<p><strong>12.2. </strong> Features, functions, and nonlinear classification <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_2_Introduction_nonlinear_classification.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/12_Nonlinear_intro/12_2_Introduction_nonlinear_classification_SLIDES.slides.html#/"> slides</a></p>
<p><strong>12.3. </strong> Features, functions, and nonlinear unsupervised learning <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_3_Introduction_nonlinear_unsupervised.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/12_Nonlinear_intro/12_3_Introduction_nonlinear_unsupervised_SLIDES.slides.html#/"> slides</a></p>
<p><strong>12.4. </strong> Automating nonlinear learning <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_4_Automating_nonlinear_learning.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/12_Nonlinear_intro/12_4_Catalogs_of_functions_SLIDES.slides.html#/"> slides</a></p>
<p><strong>12.5. </strong> Universal approximation and the bias-variance trade-off <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_5_Universal_approximation.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/12_Nonlinear_intro/12_4_Catalogs_of_functions_SLIDES.slides.html#/"> slides</a></p>
<p><strong>12.6. </strong> Validation error <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_6_Validation_error.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>12.7. </strong> Model search via Boosting <a target="_blank" class="sublink-active" href="blog_posts/12_Nonlinear_intro/12_7_Boosting.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>12.8. </strong> Model search via Regularization <a target="_blank" class="sublink-inactive" href="blog_posts/12_Nonlinear_intro/12_7_Boosting.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>12.9. </strong> Bagging and general ensembling <a target="_blank" class="sublink-inactive" href="blog_posts/12_Nonlinear_intro/12_7_Boosting.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>12.10. </strong> Testing error <a target="_blank" class="sublink-inactive" href="blog_posts/12_Nonlinear_intro/12_7_Boosting.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter13" checked/>
<label for="chapter13"><span>CHAPTER 13. Multi-layer perceptrons (MLPs) </span></label>
<div class="nine_col">
<div class="description">
<p><strong>13.1. </strong> Introduction to multi-layer perceptrons <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_1_Multi_layer_perceptrons.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/13_Multilayer_perceptrons/13_1_Multi_layer_perceptrons_SLIDES.slides.html#/"> slides</a></p>
<p><strong>13.2. </strong> The backpropagation algorithm <a target="_blank" class="sublink-inactive" href="blog_posts/13_Multilayer_perceptrons/13_1_Multi_layer_perceptrons.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/13_Multilayer_perceptrons/13_1_Multi_layer_perceptrons_SLIDES.slides.html#/"> slides</a></p>
<p><strong>13.3. </strong> Batch normalization <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_3_Batch_normalization.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>13.4. </strong> Normalized gradient descent <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_4_Normalized_gradient_descent.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>13.5. </strong> Momentum methods <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_5_Momentum_methods.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/13_Multilayer_perceptrons/13_4_Momentum_trick_SLIDES.slides.html#/"> slides</a></p>
<p><strong>13.6. </strong> Regularization <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_6_regularization.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/13_Multilayer_perceptrons/13_6_regularization_SLIDES.slides.html#/"> slides</a></p>
<p><strong>13.7. </strong> Stochastic and mini-batch gradient descent <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_7_Stochastic_and_minibatch_gradient_descent.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>13.8. </strong> General steepest descent <a target="_blank" class="sublink-active" href="blog_posts/13_Multilayer_perceptrons/13_8_General_steepest_descent.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>13.9. </strong> Early stopping <a target="_blank" class="sublink-inactive" href=""> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter14" checked/>
<label for="chapter14"><span>CHAPTER 14. Convolutional Neural Networks (CNNs) </span></label>
<div class="eight_col">
<div class="description">
<p><strong>14.1. </strong> The convolution operation <a target="_blank" class="sublink-active" href="blog_posts/14_Convolutional_networks/14_1_The_convolution_operation.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/Convolutional_Networks/14_1_The_convolution_operation_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.2. </strong> Edge histogram based features <a target="_blank" class="sublink-active" href="blog_posts/14_Convolutional_networks/14_2_Edge_histogram_based_features.html"> text</a> <a target="_blank" class="sublink-active" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.3. </strong> Convolutional neural networks <a target="_blank" class="sublink-active" href="blog_posts/14_Convolutional_networks/14_3_Convolutional_neural_network.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.4. </strong> Contrast normalization and feature scaling via ZCA sphereing <a target="_blank" class="sublink-active" href="blog_posts/14_Convolutional_networks/14_4_ZCA_Sphereing.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.5. </strong> Engineering bbject detection and recognition systems <a target="_blank" class="sublink-inactive" href="blog_posts/14_Convolutional_networks/14_2_Edge_histogram_based_features.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.6. </strong> A short list of historically important architectures <a target="_blank" class="sublink-inactive" href="blog_posts/14_Convolutional_networks/14_2_Edge_histogram_based_features.html"> text</a> <a target="_blank" class="sublink-active" href="blog_posts/14_Convolutional_networks/14_4_Deep_Convolutional_Networks.html"> slides</a></p>
<p><strong>14.7. </strong> Transfer learning <a target="_blank" class="sublink-inactive" href="blog_posts/14_Convolutional_networks/14_2_Edge_histogram_based_features.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
<p><strong>14.8. </strong> Adversarial examples and the fragility of convolutional networks <a target="_blank" class="sublink-inactive" href="blog_posts/14_Convolutional_networks/14_2_Edge_histogram_based_features.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_2_Edge_histogram_based_features_SLIDES.slides.html#/"> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter14" checked/>
<label for="chapter15"><span>CHAPTER 15. Recurrent Neural Networks (RNNs) </span></label>
<div class="eight_col">
<div class="description">
<p><strong>15.1. </strong> Introduction <a target="_blank" class="sublink-active" href="blog_posts/15_Recurrent_Networks/15_1_Introduction.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/14_1_The_convolution_operation_SLIDES.slides.html#/"> slides</a></p>
<p><strong>15.2. </strong> Fixed Order Dynamic Systems <a target="_blank" class="sublink-active" href="blog_posts/15_Recurrent_Networks/15_2_Fixed_Order.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.3. </strong> Fixed order generative models <a target="_blank" class="sublink-active" href="blog_posts/15_Recurrent_Networks/15_3_Recurrence_Relations.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.4. </strong> Variable order dynamic systems <a target="_blank" class="sublink-active" href="blog_posts/15_Recurrent_Networks/15_4_Variable_Order.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.5. </strong> Autoregressive modeling <a target="_blank" class="sublink-inactive" href="blog_posts/15_Recurrent_Networks/15_2_Fixed_Order.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.6. </strong> Recurrent networks <a target="_blank" class="sublink-inactive" href="blog_posts/15_Recurrent_Networks/15_2_Memoryless.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.7. </strong> Optimization tricks for recurrent networks <a target="_blank" class="sublink-inactive" href="blog_posts/15_Recurrent_Networks/15_2_Memoryless.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
<p><strong>15.8. </strong> Advanced architectures <a target="_blank" class="sublink-inactive" href="blog_posts/15_Recurrent_Networks/15_2_Memoryless.html"> text</a> <a target="_blank" class="sublink-inactive" href="presentations/Convolutional_Networks/.slides.html#/"> slides</a></p>
</div>
</div>
</div>
<div class="chapters-container">
<input type="checkbox" name="chapters" id="chapter18" checked/>
<label for="chapter18"><span>CHAPTER 18. Reinforcement learning foundations</span></label>
<div class="four_col">
<div class="description">
<p><strong>18.1. </strong> Fundamentals of Reinforcement Learning <a target="_blank" class="sublink-active" href="blog_posts/18_Reinforcement_Learning_Foundations/18_1_Fundamentals_of_reinforcement_learning.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>18.2. </strong> Q-Learning <a target="_blank" class="sublink-active" href="blog_posts/18_Reinforcement_Learning_Foundations/18_2_Q_learning.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>18.3. </strong> Q-Learning enhancements <a target="_blank" class="sublink-active" href="blog_posts/18_Reinforcement_Learning_Foundations/18_3_Q_learning_enhancements.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
<p><strong>18.4. </strong> On generalizability of Reinforcement Learning <a target="_blank" class="sublink-active" href="blog_posts/18_Reinforcement_Learning_Foundations/18_4_On_generalizability_of_reinforcement_learning.html"> text</a> <a target="_blank" class="sublink-inactive" href=""> slides</a></p>
</div>
</div>
</div>
<br/><br/><br/><br/><br/><br/><br/><br/><br/><br/><br/><br/><br/><br/>
<!-- uncomment subscription
<!- - subscription button - ->
<form action="https://formspree.io/[email protected]" method="POST">
<input type="email" name="email" placeholder="Enter your email to get notified when new posts are published" onfocus="this.placeholder=''" onblur="this.placeholder='Enter your email to get notified when new posts are published'" autocomplete="off">
<button type="submit" value="Send">Subscribe <i class="fa fa-envelope-o"></i></button>
</form>
-->
<br/><br/><br/><br/><br/><br/><br/><br/><br/><br/>
<script>
$(document).ready(function(){
$('.hover').hover(function(){
$(this).addClass('flip');
},function(){
$(this).removeClass('flip');
});
});
</script>
</body>
</html>