-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathgradient.drawio
150 lines (150 loc) · 14.8 KB
/
gradient.drawio
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
<mxfile host="app.diagrams.net" modified="2023-06-15T08:38:21.409Z" agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/114.0.0.0 Safari/537.36 Edg/114.0.1823.43" etag="SxOy9bDZP2F8-XSaGcwG" version="21.3.8" type="github">
<diagram name="Page-1" id="iOJ7nJUNVYEBUDQMrWRL">
<mxGraphModel dx="1100" dy="634" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="1169" pageHeight="827" math="0" shadow="0">
<root>
<mxCell id="0" />
<mxCell id="1" parent="0" />
<mxCell id="CzOLXgxaiCzq6wpnYxIb-24" value="Convergent and termination" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="570" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-64" value="Initial model training" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="540" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-65" value="Continous learning" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="150" y="540" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-66" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-64" target="CzOLXgxaiCzq6wpnYxIb-65">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-67" value="Data processing" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="290" y="540" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-68" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-65" target="CzOLXgxaiCzq6wpnYxIb-67">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-97" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-70" target="CzOLXgxaiCzq6wpnYxIb-75">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-70" value="Model update" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="430" y="540" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-71" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-67" target="CzOLXgxaiCzq6wpnYxIb-70">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-72" value="Benefit of incremental learning for dynamic adaptive powertrain control<br>1. Real time adaptation<br>2. knowledge preservation<br>3. Efficiency and scalabiltiy<br>4. Fault tolerance" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="690" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-73" value="Challenges incremental learning for dynamic adaptive powertrain control<br>1. Catastrophic forgetting<br>2. Data quality and bias&nbsp;<br>3. model complexity<br>" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#b0e3e6;strokeColor=#0e8088;" vertex="1" parent="1">
<mxGeometry x="294" y="690" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-74" value="Incremental learning for dynamic adaptation" style="rounded=1;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="500" width="250" height="20" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-75" value="Convergent and termination" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="600" y="540" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-94" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.558;exitY=0.983;exitDx=0;exitDy=0;exitPerimeter=0;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-75" target="CzOLXgxaiCzq6wpnYxIb-93">
<mxGeometry relative="1" as="geometry">
<mxPoint x="650" y="570" as="sourcePoint" />
<Array as="points">
<mxPoint x="667" y="650" />
</Array>
</mxGeometry>
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-95" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;entryX=0.417;entryY=1;entryDx=0;entryDy=0;entryPerimeter=0;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-93" target="CzOLXgxaiCzq6wpnYxIb-70">
<mxGeometry relative="1" as="geometry">
<Array as="points">
<mxPoint x="480" y="650" />
</Array>
</mxGeometry>
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-93" value="Adaptation and Feedback loop" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="510" y="620" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-1" value="Initiatlization" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-2" value="Data Acquisition" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="150" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-3" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-1" target="CzOLXgxaiCzq6wpnYxIb-2">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-4" value="Prediction and error calculations" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="290" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-5" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-2" target="CzOLXgxaiCzq6wpnYxIb-4">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-6" value="Parameter Update" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="430" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-7" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-4" target="CzOLXgxaiCzq6wpnYxIb-6">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-8" value="Adaptation and learning" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="570" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-9" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-6" target="CzOLXgxaiCzq6wpnYxIb-8">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-10" value="Convergent and termination" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="710" y="40" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-11" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-8" target="CzOLXgxaiCzq6wpnYxIb-10">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-12" value="Benefit of&nbsp;gradient descent for adaptive powertrain control<br>1. Real time adaptation<br>2. Efficiency and scalability<br>3. Customization and fine tuning<br>4. Dynamic optimization" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="120" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-13" value="Challenges of gradient descent for adaptive powertrain control<br><br>1. Concept Drift<br>2. Data Quality and noise&nbsp;<br>3. Model complexity&nbsp;" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#b1ddf0;strokeColor=#10739e;" vertex="1" parent="1">
<mxGeometry x="300" y="120" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-27" value="Gradient Descent algorithm" style="rounded=1;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="10" width="150" height="20" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-15" value="Initiatlization" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-17" value="Data Acquisition" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="150" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-14" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-15" target="CzOLXgxaiCzq6wpnYxIb-17">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-19" value="Data sampling" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="290" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-16" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-17" target="CzOLXgxaiCzq6wpnYxIb-19">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-20" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-21">
<mxGeometry relative="1" as="geometry">
<mxPoint x="570" y="310" as="targetPoint" />
</mxGeometry>
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-21" value="Gradient calculation" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="430" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-18" value="" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#fad9d5;strokeColor=#ae4132;" edge="1" parent="1" source="CzOLXgxaiCzq6wpnYxIb-19" target="CzOLXgxaiCzq6wpnYxIb-21">
<mxGeometry relative="1" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-25" value="Benefit of stochastic gradient descent for adaptive powertrain control<br>1. Efficiency and scalability<br>2. Real time adaptation<br>3. Regularization and generalization<br>4. Dynamic optimization" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="360" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-26" value="Challenges stochastic of gradient descent for adaptive powertrain control<br><br>1. Learning Rate selection<br>2. Noise and outliers&nbsp;<br>3. Local optima" style="rounded=0;whiteSpace=wrap;html=1;align=left;fillColor=#b0e3e6;strokeColor=#0e8088;" vertex="1" parent="1">
<mxGeometry x="300" y="360" width="240" height="100" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-28" value="Stochastic Gradient Descent algorithm&nbsp;" style="rounded=1;whiteSpace=wrap;html=1;align=left;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="10" y="240" width="220" height="20" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-31" value="Convergent and termination" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="570" y="280" width="120" height="60" as="geometry" />
</mxCell>
<mxCell id="CzOLXgxaiCzq6wpnYxIb-32" value="Convergent and termination" style="whiteSpace=wrap;html=1;rounded=0;fillColor=#fad9d5;strokeColor=#ae4132;" vertex="1" parent="1">
<mxGeometry x="570" y="280" width="120" height="60" as="geometry" />
</mxCell>
</root>
</mxGraphModel>
</diagram>
</mxfile>