0.25
0.5
0.75
1.25
1.5
1.75
2
Stacks of Restricted Boltzmann Machines
Published on Sep 13, 201510460 Views
Related categories
Chapter list
Stacks of Restricted Boltzmann Machines - 100:00
Stacks of Restricted Boltzmann Machines - 201:24
Learning Feature Hierarchy - 101:27
Learning Feature Hierarchy - 202:30
Restricted Boltzmann Machines (RBMs) - 102:58
Restricted Boltzmann Machines (RBMs) - 204:01
Conditional Probabilities (RBM with binary-valued input data)05:23
RBMs with real-valued input data - 106:50
RBMs with real-valued input data - 206:53
Inference07:28
Training RBMs - 108:30
Training RBMs - 208:53
Contrastive Divergence09:02
Maximum likelihood learning for RBM09:04
Contrastive divergence to learn RBM09:05
Update rule: Putting together09:06
Other ways of training RBMs09:59
Variants of RBMs10:26
Sparse RBM / DBN10:30
Modeling handwritten digits12:09
3-way factorized RBM12:59
Generating natural image patches14:28
Stacking of RBMs as Deep Belief Networks15:26
Deep Belief Networks (DBNs) - 115:35
Deep Belief Networks (DBNs) - 217:51
DBN structure19:27
DBN Greedy training - 125:27
DBN Greedy training - 226:12
DBN Greedy training - 328:12
Why greedy training works? - 128:43
Why greedy training works? - 229:06
Why greedy training works? - 330:07
Derivation - 131:04
Derivation - 256:15
Theoretical Justification (summary)57:09
DBN and supervised fine-tuning57:31
Generative fine-tuning via Up-down algorithm01:02:34
A model for digit recognition01:04:38
Generating sample from a DBN01:05:58
Generating samples from DBN01:06:02
Result for supervised fine-tuning on MNIST01:06:03
More details on up-down algorithm01:06:03
Stacking of RBMs as Deep Neural Networks01:06:15
Using Stacks of RBMs as Neural Networks01:06:17
DBN for classification01:06:58
Stacks of RBMs as deep autoencoders - 101:07:26
Stacks of RBMs as deep autoencoders - 201:08:18
Learning similarity metric01:09:00
Learning feature hierarchy for images01:09:45
Speech recognition using DBNs01:11:14
Other applications01:11:36
Remarks01:12:22
DBN inference example: Image Inpainting for Facial Expression Recognition01:12:32
Another Example of a hybrid graphical model01:14:32
Learning Output Representations01:19:54
Structured output prediction: Examples01:20:12
Combining Global and Local Consistencies for Structured Output Prediction01:20:55
CRF: baseline segmentation labeler01:21:09
Output of CRF = Visible of RBM01:22:47
Formulation01:23:28
H learns shape patterns01:23:42
Generated Samples from RBM prior01:24:25
Prediction: mean-field approximation01:24:48
Prediction: mean-field approximation01:25:21
Experimental results01:26:33
Quantitative Evaluation on Labeling01:27:48