en-de
en-es
en-fr
en-pt
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
On-line learning algorithms: theory and practice
Published on Dec 14, 200711267 Views
Related categories
Chapter list
On-Line Learning00:00
Summary00:37
Summary02:54
On-line classification02:55
Linear classifiers04:22
On-line learning protocol - 106:46
On-line learning protocol - 206:58
On-line learning protocol - 307:58
Remarks10:18
Summary14:17
Perceptron algorithm - 114:18
Perceptron algorithm - 215:24
Linear separability - 117:34
Linear separability - 218:19
Relative loss bound - 119:09
Relative loss bound - 220:55
Relative loss bound - 321:55
Norm of the separator23:30
Analysis of Perceptron25:41
Analysis: When a mistake occurs - 127:18
Analysis: When a mistake occurs - 228:49
When a mistake occurs (cont.) - 129:39
When a mistake occurs (cont.) - 230:19
When a mistake occurs (cont.) - 330:28
The relative mistake bound - 130:51
The relative mistake bound - 231:23
Summary34:07
Aggressive updates: Hildreth algorithm (1957) - 134:13
Aggressive updates: Hildreth algorithm (1957) - 235:01
Aggressive updates: Hildreth algorithm (1957) - 335:59
Aggressive updates: Hildreth algorithm (1957) - 436:36
Aggressive updates: Hildreth algorithm (1957) - 537:09
Aggressive updates: Hildreth algorithm (1957) - 637:41
Aggressive updates: Hildreth algorithm (1957) - 738:26
Analysis for linearly separable streams - 139:07
Analysis for linearly separable streams - 239:29
Analysis for linearly separable streams - 340:00
Analysis (cont.)46:41
The cone of consistent hyperplanes - 147:18
The cone of consistent hyperplanes - 349:05
The cone of consistent hyperplanes - 249:07
The cone of consistent hyperplanes - 450:08
The cone of consistent hyperplanes - 550:17
The cone of consistent hyperplanes - 650:37
The cone of consistent hyperplanes - 751:11
Mistake bounds for various updates - 151:47
Mistake bounds for various updates - 252:03
Mistake bounds for various updates - 355:39
Summary55:55
Aggressive updates for nonseparable streams - 156:05
Aggressive updates for nonseparable streams - 257:19
Aggressive updates for nonseparable streams - 358:08
Aggressive updates for nonseparable streams - 459:05
Aggressive updates for nonseparable streams - 559:12
SVM and passive-aggressive - 101:00:15
SVM and passive-aggressive - 201:01:15
SVM and passive-aggressive - 301:02:06
SVM and passive-aggressive (cont.) - 101:02:56
SVM and passive-aggressive (cont.) - 201:03:34
SVM and passive-aggressive (cont.) - 301:04:08
SVM and passive-aggressive (cont.) - 401:04:43
SVM and passive-aggressive (cont.) - 501:04:57
Mistake bounds for PA-I - 101:05:33
Mistake bounds for PA-I - 201:06:25
Mistake bounds for PA-I - 301:06:53
Proof of mistake bound for PA-I - 101:07:37
Proof of mistake bound for PA-I - 201:07:41
Proof of mistake bound for PA-I - 301:07:53
Proof of mistake bound for PA-I - 401:08:50
Summary01:11:46
On-line learning with kernels - 101:12:11
On-line learning with kernels - 201:12:25
On-line learning with kernels - 301:12:31
On-line learning with kernels - 401:14:19
Kernel Perceptron - 101:15:32
Kernel Perceptron - 201:15:35
Kernel Perceptron - 301:15:36
Kernel Perceptron - 401:15:54
Kernel Perceptron - 501:15:56
Kernel Perceptron - 601:17:08
Memory bounded learning - 101:17:40
Memory bounded learning - 201:18:49
Memory bounded learning - 301:21:03
Memory bounded learning - 401:21:21
A randomized perceptron - 101:21:40
A randomized perceptron - 201:21:53
A randomized perceptron - 301:21:58
A randomized perceptron - 401:22:06
A randomized perceptron - 501:22:08
A randomized perceptron - 601:22:10
A randomized perceptron - 701:22:23
A randomized perceptron - 801:22:59
Empirical performance - stationary01:24:16
Empirical performance - nonstationary01:26:49
Empirical performance 2nd order - nonstationary01:27:53