50:, though it has been used in other domains too. The salient features of the model include the fact that it scales well to high-dimensional images and is translation-invariant.
146:
207:
65:. Depending on whether the network is to be used for discrimination or generative tasks, it is then "fine tuned" or trained with either
178:
57:
to reduce the dimensions in higher layers in the network. Training of the network involves a pre-training stage accomplished in a
93:
212:
35:
32:
28:
111:
139:
87:"Convolutional Deep Belief Networks for Scalable Unsupervised Learning of Hierarchical Representations"
62:
140:"Unsupervised feature learning for audio classification using convolutional deep belief networks"
47:
58:
43:
39:
17:
124:
66:
171:
86:
201:
25:
54:
69:
or the up–down algorithm (contrastive–divergence), respectively.
38:
stacked together. Alternatively, it is a hierarchical
138:
Lee, Honglak; Yan
Largman; Peter Pham; Andrew Y. Ng.
42:for deep learning, which is highly effective in
8:
85:Lee, Honglak; Grosse, Ranganath; Andrew Ng.
53:CDBNs use the technique of probabilistic
77:
120:
109:
7:
172:"Convolutional Deep Belief Networks"
61:layer-wise manner, similar to other
14:
22:convolutional deep belief network
184:from the original on 2014-04-07
152:from the original on 2023-01-28
99:from the original on 2014-04-07
31:composed of multiple layers of
1:
36:restricted Boltzmann machines
229:
208:Artificial neural networks
29:artificial neural network
119:Cite journal requires
213:Probabilistic models
170:Coviello, Emanuele.
63:deep belief networks
24:(CDBN) is a type of
48:object recognition
220:
193:
192:
190:
189:
183:
176:
167:
161:
160:
158:
157:
151:
144:
135:
129:
128:
122:
117:
115:
107:
105:
104:
98:
91:
82:
67:back-propagation
44:image processing
40:generative model
18:computer science
228:
227:
223:
222:
221:
219:
218:
217:
198:
197:
196:
187:
185:
181:
174:
169:
168:
164:
155:
153:
149:
142:
137:
136:
132:
118:
108:
102:
100:
96:
89:
84:
83:
79:
75:
12:
11:
5:
226:
224:
216:
215:
210:
200:
199:
195:
194:
162:
130:
121:|journal=
76:
74:
71:
13:
10:
9:
6:
4:
3:
2:
225:
214:
211:
209:
206:
205:
203:
180:
173:
166:
163:
148:
141:
134:
131:
126:
113:
95:
88:
81:
78:
72:
70:
68:
64:
60:
56:
51:
49:
45:
41:
37:
34:
33:convolutional
30:
27:
23:
19:
186:. Retrieved
165:
154:. Retrieved
133:
112:cite journal
101:. Retrieved
80:
52:
21:
15:
55:max-pooling
202:Categories
188:2014-04-01
156:2019-08-25
103:2014-04-01
73:References
179:Archived
147:Archived
94:Archived
59:greedy
182:(PDF)
175:(PDF)
150:(PDF)
143:(PDF)
97:(PDF)
90:(PDF)
125:help
46:and
26:deep
20:, a
16:In
204::
177:.
145:.
116::
114:}}
110:{{
92:.
191:.
159:.
127:)
123:(
106:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.