forked from mini-pw/2021L-WB-Book
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path4-5-RMDLAdv.Rmd
266 lines (238 loc) · 8.59 KB
/
4-5-RMDLAdv.Rmd
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
## Title
*Authors: Filip Chrzuszcz, Szymon Rećko, Mateusz Sperkowski (Warsaw University of Technology)*
### Title, Authors, Abstract, Keywords
### Introduction
### Related Literature
### Methods
### Result
Skrot do pierwszego projektu: Despite lower results than in the first paper, in most datasets we still achieved better results than the baselines paper attempted to beat. The ones that we weren't able to reproduce where either limits of processing power, or could be assigned to effect off randomness which is basis od this paper. The authors unfortunately didn't include their randomness results, therefore their exact calculations aren't reproducible.
### Discussion
### Conclusion
### References
### Random Multimodel Deep Learning for Classification
Results
<style type="text/css">
.tg {border-collapse:collapse;border-spacing:0;}
.tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-c3ow{border-color:inherit;text-align:center;vertical-align:top}
.tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top}
</style>
<table class="tg">
<thead>
<tr>
<th class="tg-0pky" colspan="2" rowspan="2"></th>
<th class="tg-c3ow" colspan="8">Dataset</th>
</tr>
<tr>
<td class="tg-c3ow" colspan="2">WOS-5736</td>
<td class="tg-c3ow" colspan="2">WOS-11967</td>
<td class="tg-c3ow" colspan="2">WOS-46985</td>
<td class="tg-c3ow" colspan="2">Reuters-21578</td>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-c3ow" colspan="2">Score Source<br></td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
</tr>
<tr>
<td class="tg-c3ow" rowspan="4">RMDL</td>
<td class="tg-0pky">3 RDLs<br></td>
<td class="tg-0pky">90.86</td>
<td class="tg-0pky">89.37</td>
<td class="tg-0pky">87.39</td>
<td class="tg-0pky">84.25</td>
<td class="tg-0pky">78.39</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">89.10</td>
<td class="tg-0pky">87.64</td>
</tr>
<tr>
<td class="tg-0pky">9 RDLs</td>
<td class="tg-0pky">92.60</td>
<td class="tg-0pky">89.28</td>
<td class="tg-0pky">90.65</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">81.92</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">90.36</td>
<td class="tg-0pky">89.83</td>
</tr>
<tr>
<td class="tg-0pky">15 RDLs</td>
<td class="tg-0pky">92.66</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">91.01</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">81.86</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">89.91</td>
<td class="tg-0pky">---</td>
</tr>
<tr>
<td class="tg-0pky">30 RDLs</td>
<td class="tg-0pky">93.57</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">91.59</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">82.42</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">90.69</td>
<td class="tg-0pky">---</td>
</tr>
</tbody>
</table>
<center>Table 1</center>
<br>
#### Reuters-21578
Paper's Plots | Our Reproduction
--- | ---
<img src="images/4-5-ReutersPaperTrain.png" width="425" height="310"/> | <img src="images/4-5-ReutersReproductionTrain.png" width="425"/>
<img src="images/4-5-ReutersPaperTest.png" width="425" height="305"/> | <img src="images/4-5-ReutersReproductionTest.png" width="425"/>
<center>Figure 1</center>
#### WOS-5736
Paper's Plots | Our Reproduction
--- | ---
<img src="images/4-5-WOS5736PaperTrain.png" width="425" height="290"/> | <img src="images/4-5-WOS5736ReproductionTrain.png" width="425"/>
<img src="images/4-5-WOS5736PaperTest.png" width="425" height="285"/> | <img src="images/4-5-WOS5736ReproductionTest.png" width="425"/>
<center>Figure 2</center>
<style type="text/css">
.tg {border-collapse:collapse;border-spacing:0;}
.tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-9wq8{border-color:inherit;text-align:center;vertical-align:middle}
.tg .tg-c3ow{border-color:inherit;text-align:center;vertical-align:top}
.tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top}
</style>
<table class="tg">
<thead>
<tr>
<th class="tg-0pky" colspan="2" rowspan="2"></th>
<th class="tg-c3ow" colspan="4">Dataset</th>
</tr>
<tr>
<td class="tg-c3ow" colspan="2">IMDB</td>
<td class="tg-c3ow" colspan="2">20NewsGroup<br></td>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-c3ow" colspan="2">Score Source<br></td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
</tr>
<tr>
<td class="tg-9wq8" rowspan="3">RMDL</td>
<td class="tg-0pky">3 RDLs</td>
<td class="tg-0pky">89.91</td>
<td class="tg-0pky">88.49</td>
<td class="tg-0pky">86.73</td>
<td class="tg-0pky">---</td>
</tr>
<tr>
<td class="tg-0pky">9 RDLs</td>
<td class="tg-0pky">90.13</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">87.62</td>
<td class="tg-0pky">---</td>
</tr>
<tr>
<td class="tg-0pky">15 RDLs</td>
<td class="tg-0pky">90.79</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">87.91</td>
<td class="tg-0pky">---</td>
</tr>
</tbody>
</table>
<center>Table 2</center>
<br>
ERROR RATE 1-Accuracy
<style type="text/css">
.tg {border-collapse:collapse;border-spacing:0;}
.tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-9wq8{border-color:inherit;text-align:center;vertical-align:middle}
.tg .tg-c3ow{border-color:inherit;text-align:center;vertical-align:top}
.tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top}
</style>
<table class="tg">
<thead>
<tr>
<th class="tg-0pky" colspan="2" rowspan="2"></th>
<th class="tg-c3ow" colspan="4">Dataset</th>
</tr>
<tr>
<td class="tg-c3ow" colspan="2">MNIST</td>
<td class="tg-c3ow" colspan="2">CIFAR-10<br></td>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-c3ow" colspan="2">Score Source<br></td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
<td class="tg-c3ow">Paper</td>
<td class="tg-c3ow">Repr.</td>
</tr>
<tr>
<td class="tg-9wq8" rowspan="4">RMDL</td>
<td class="tg-0pky">3 RDLs<br></td>
<td class="tg-0pky">0.51<br></td>
<td class="tg-0pky">0.55<br></td>
<td class="tg-0pky">9.89</td>
<td class="tg-0pky">38.23</td>
</tr>
<tr>
<td class="tg-0pky">9 RDLs</td>
<td class="tg-0pky">0.41</td>
<td class="tg-0pky">0.65</td>
<td class="tg-0pky">9.1</td>
<td class="tg-0pky">36.91</td>
</tr>
<tr>
<td class="tg-0pky">15 RDLs</td>
<td class="tg-0pky">0.21</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">8.74<br></td>
<td class="tg-0pky">---</td>
</tr>
<tr>
<td class="tg-0pky">30 RDLs</td>
<td class="tg-0pky">0.18</td>
<td class="tg-0pky">---</td>
<td class="tg-0pky">8.79</td>
<td class="tg-0pky">---</td>
</tr>
</tbody>
</table>
<center>Table 3</center>
#### CIFAR 10
Paper's Plots | Our Reproduction
--- | ---
<img src="images/4-5-Cifar10PaperTrain.png" width="425" height="285"/> | <img src="images/4-5-Cifar10ReproductionTrain.png" width="425"/>
<img src="images/4-5-Cifar10PaperTest.png" width="425" height="300"/> | <img src="images/4-5-Cifar10ReproductionTest.png" width="425"/>
<center>Figure 3</center>
#### MNIST
Paper's Plots | Our Reproduction
--- | ---
<img src="images/4-5-MNISTPaperTrain.png" width="425" height="300"/> | <img src="images/4-5-MNISTReproductionTrain.png" width="425"/>
<img src="images/4-5-MNISTPaperTest.png" width="425" height="300"/> | <img src="images/4-5-MNISTReproductionTest.png" width="425"/>
<center>Figure 4</center>
### Adversarial Attacks Against Medical Deep Learning Systems