Perfect learning in neural networks: on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Abschlussarbeit Buch |
| Sprache: | German |
| Veröffentlicht: |
1995
|
| Schlagworte: | |
| Online-Zugang: | Inhaltsverzeichnis |
| Beschreibung: | XI, 162 S. Ill., graph. Darst. |
Internformat
MARC
| LEADER | 00000nam a2200000 c 4500 | ||
|---|---|---|---|
| 001 | BV010433373 | ||
| 003 | DE-604 | ||
| 005 | 19960305 | ||
| 007 | t | ||
| 008 | 951004s1995 gw ad|| m||| 00||| ger d | ||
| 016 | 7 | |a 945298056 |2 DE-101 | |
| 035 | |a (OCoLC)64527236 | ||
| 035 | |a (DE-599)BVBBV010433373 | ||
| 040 | |a DE-604 |b ger |e rakddb | ||
| 041 | 0 | |a ger | |
| 044 | |a gw |c DE | ||
| 049 | |a DE-29T |a DE-703 |a DE-83 |a DE-11 |a DE-188 | ||
| 100 | 1 | |a Lange, Rupert |e Verfasser |4 aut | |
| 245 | 1 | 0 | |a Perfect learning in neural networks |b on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules |c vorgelegt von Rupert Lange |
| 264 | 1 | |c 1995 | |
| 300 | |a XI, 162 S. |b Ill., graph. Darst. | ||
| 336 | |b txt |2 rdacontent | ||
| 337 | |b n |2 rdamedia | ||
| 338 | |b nc |2 rdacarrier | ||
| 502 | |a Heidelberg, Univ., Diss., 1995 | ||
| 650 | 0 | 7 | |a Mehrschichten-Perzeptron |0 (DE-588)4354626-2 |2 gnd |9 rswk-swf |
| 650 | 0 | 7 | |a Backpropagation-Algorithmus |0 (DE-588)4354627-4 |2 gnd |9 rswk-swf |
| 650 | 0 | 7 | |a Maschinelles Lernen |0 (DE-588)4193754-5 |2 gnd |9 rswk-swf |
| 655 | 7 | |0 (DE-588)4113937-9 |a Hochschulschrift |2 gnd-content | |
| 689 | 0 | 0 | |a Mehrschichten-Perzeptron |0 (DE-588)4354626-2 |D s |
| 689 | 0 | 1 | |a Maschinelles Lernen |0 (DE-588)4193754-5 |D s |
| 689 | 0 | 2 | |a Backpropagation-Algorithmus |0 (DE-588)4354627-4 |D s |
| 689 | 0 | |5 DE-604 | |
| 856 | 4 | 2 | |m DNB Datenaustausch |q application/pdf |u http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=006952444&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA |3 Inhaltsverzeichnis |
| 943 | 1 | |a oai:aleph.bib-bvb.de:BVB01-006952444 | |
Datensatz im Suchindex
| _version_ | 1807502478315356160 |
|---|---|
| adam_text |
CONTENTS
LIST
OF
FIGURES
V
LIST
OF
TABLES
VII
ABBREVIATIONS
IX
SYMBOLS
XI
1
INTRODUCTION
1
1.1
ARTIFICIAL
NEURAL
NETWORKS
.
1
1.2
UPDATE
ALGORITHM
.
2
1.3
LEARNING
ALGORITHM
.
4
1.4
GENERALIZATION
.
4
1.5
NEURAL
NETWORKS
'
PROMISE
.
5
1.6
MOTIVATION
AND
STRUCTURE
OF
THESIS
.
11
2
THE
TEACHER-PUPIL
PROBLEM
13
2.1
INTRODUCTION
.
13
2.1.1
LEARNING
A
RULE
-
TERMINOLOGY
.
13
2.1.2
THE
TEACHER-PUPIL
PROBLEM
.
15
2.1.3
THEORY
.
16
2.2
SIMPLE
PERCEPTRONS
.
20
2.2.1
THRESHOLD
PERCEPTRON
.
21
2.2.2
LINEAR
PERCEPTRON
.
25
2.3
COMMITTEE
MACHINE
.
29
2.3.1
TREE
COMMITTEE
MACHINE
.
30
2.3.2
FULLY-CONNECTED
COMMITTEE
MACHINE
.
31
2.4
SUMMARY
.
33
3
STANDARD
BACKPROPAGATION
39
3.1
INTRODUCTION
.
39
3.2
EXPERIMENTAL
SETTING
.
43
3.3
RESULTS
FOR
N-N-N
ARCHITECTURES
.
45
3.3.1
TRAINING
.
46
3.3.2
GENERALIZATION
.
49
3.3.3
COMPUTATIONAL
COMPLEXITY
.
53
3.4
MODIFICATIONS
.
53
II
CONTENTS
3.4.1
ORDER
OF
EXAMPLE
PRESENTATION
.
54
3.4.2
SEMI-LINEAR
TRANSFER
FUNCTION
.
55
3.4.3
16-16-N
.
57
3.4.4
N-N-N-N
ARCHITECTURES
.
58
3.4.5
APPLICATION-TRAINED
TEACHER
.
58
3.5
SUMMARY
.
60
4
NORMALIZED
BACKPROPAGATION
65
4.1
INTRODUCTION
.
65
4.2
STANDARD
BACKPROPAGATION
.
66
4.2.1
FORWARD
PASS
.
66
4.2.2
BACKWARD
PASS
.
68
4.2.3
ONLINE
TRAINING
ON
SQUARE
ERROR
.
68
4.3
SCALING
ISSUES
.
69
4.3.1
PRELIMINARIES
.
69
4.3.2
ORDER
OF
FIELDS
AND
STATES
.
70
4.3.3
INITIAL
ERROR
.
73
4.3.4
NONLINEARITY
.
75
4.3.5
LAYER
CHARACTERISTIC
.
77
4.3.6
ORDER
OF
WEIGHT
CHANGES
.
79
4.3.7
WEIGHT
CHANGE
SCENARIOS
AND
TRAINING
TIME
.
80
4.4
NORMALIZED
BACKPROPAGATION
.
82
4.4.1
FORWARD
PASS
.
83
4.4.2
BACKWARD
PASS
.
83
4.4.3
INTERPRETATION
.
84
4.5
SAMPLE
COMPLEXITY
.
85
4.6
CRITICAL
STEP
SIZE
.
92
4.6.1
STANDARD
BACKPROPAGATION
.
93
4.6.2
NORMALIZED
BACKPROPAGATION
.
94
4.7
SUMMARY
.
97
5
MONNET
99
5.1
INTRODUCTION
.
99
5.2
DESIGN
.
100
5.2.1
LOGICAL
STRUCTURE
.
101
5.2.2
EFFICIENCY
.
103
5.2.3
USER
INTERFACE
.
103
5.3
IMPLEMENTATION
.
104
5.3.1
OBJECT
ORIENTED
ENVIRONMENT
.
104
5.3.2
KERNEL
.
105
5.3.3
OBJECTS
.
105
5.3.4
NODES
.
106
5.3.5
LINKS
.
107
5.4
CASE
STUDY
.
108
5.4.1
TRAIN
AND
TEST
NN
.
108
5.4.2
EXTENSION
1:
INPUT
DISTURBANCE
.
110
CONTENTS
III
5.4.3
EXTENSION
2:
WEIGHT
DISTURBANCE
.
ILL
5.5
SUMMARY
.
113
6
USING
MONNET
115
6.1
INTRODUCTION
.
115
6.2
DESIGN
.
116
6.2.1
PARAMETERS
.
116
6.2.2
NEURAL
NETWORK
NODE
.
118
6.2.3
TRAINER
NODE
.
123
6.2.4
TESTER
NODE
.
124
6.2.5
OTHER
MODULES
.
126
6.3
MONNET
NETWORK
.
126
6.3.1
TRANSFER
FUNCTION
AND
ITS
DERIVATIVE
.
128
6.3.2
GENERATION
OF
TRAINING
AND
TESTING
SETS
.
129
6.3.3
TRAINING
THE
PUPIL
NEURAL
NETWORK
.
132
6.3.4
TEST
RUN
.
134
6.4
ENVIRONMENT
.
135
6.4.1
RUNNING
THE
SIMULATIONS
.
135
6.4.2
POSTPROCESSING
.
136
6.4.3
COLLECTING
AND
DISPLAYING
RESULTS
.
137
6.5
SUMMARY
.
138
7
CONCLUSIONS
139
A
EXPERIMENT
CODE
AND
PROTOCOL
143
A.L
CODE
.
143
A.1.1
NODES
.
144
A.L.
2
LINKS
.
145
A.2
PROTOCOL
.
146
A.2.1
PARAMETERS
.
146
A.
2.2
TRANSFER
FUNCTION
.
147
A.2.
3
TRAINING
AND
TESTING
SET
.
149
A.2.4
TRAINING
THE
PUPIL
.
151
B
MONNET
ENVIRONMENT
153
B.L
TABLE
OF
PROGRAMS
.
153
BIBLIOGRAPHY
155 |
| any_adam_object | 1 |
| author | Lange, Rupert |
| author_facet | Lange, Rupert |
| author_role | aut |
| author_sort | Lange, Rupert |
| author_variant | r l rl |
| building | Verbundindex |
| bvnumber | BV010433373 |
| ctrlnum | (OCoLC)64527236 (DE-599)BVBBV010433373 |
| format | Thesis Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>00000nam a2200000 c 4500</leader><controlfield tag="001">BV010433373</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">19960305</controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">951004s1995 gw ad|| m||| 00||| ger d</controlfield><datafield tag="016" ind1="7" ind2=" "><subfield code="a">945298056</subfield><subfield code="2">DE-101</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)64527236</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV010433373</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rakddb</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">ger</subfield></datafield><datafield tag="044" ind1=" " ind2=" "><subfield code="a">gw</subfield><subfield code="c">DE</subfield></datafield><datafield tag="049" ind1=" " ind2=" "><subfield code="a">DE-29T</subfield><subfield code="a">DE-703</subfield><subfield code="a">DE-83</subfield><subfield code="a">DE-11</subfield><subfield code="a">DE-188</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Lange, Rupert</subfield><subfield code="e">Verfasser</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Perfect learning in neural networks</subfield><subfield code="b">on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules</subfield><subfield code="c">vorgelegt von Rupert Lange</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">1995</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">XI, 162 S.</subfield><subfield code="b">Ill., graph. Darst.</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="502" ind1=" " ind2=" "><subfield code="a">Heidelberg, Univ., Diss., 1995</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Mehrschichten-Perzeptron</subfield><subfield code="0">(DE-588)4354626-2</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Backpropagation-Algorithmus</subfield><subfield code="0">(DE-588)4354627-4</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Maschinelles Lernen</subfield><subfield code="0">(DE-588)4193754-5</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="655" ind1=" " ind2="7"><subfield code="0">(DE-588)4113937-9</subfield><subfield code="a">Hochschulschrift</subfield><subfield code="2">gnd-content</subfield></datafield><datafield tag="689" ind1="0" ind2="0"><subfield code="a">Mehrschichten-Perzeptron</subfield><subfield code="0">(DE-588)4354626-2</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2="1"><subfield code="a">Maschinelles Lernen</subfield><subfield code="0">(DE-588)4193754-5</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2="2"><subfield code="a">Backpropagation-Algorithmus</subfield><subfield code="0">(DE-588)4354627-4</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2=" "><subfield code="5">DE-604</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="m">DNB Datenaustausch</subfield><subfield code="q">application/pdf</subfield><subfield code="u">http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=006952444&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA</subfield><subfield code="3">Inhaltsverzeichnis</subfield></datafield><datafield tag="943" ind1="1" ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-006952444</subfield></datafield></record></collection> |
| genre | (DE-588)4113937-9 Hochschulschrift gnd-content |
| genre_facet | Hochschulschrift |
| id | DE-604.BV010433373 |
| illustrated | Illustrated |
| indexdate | 2024-08-16T00:38:10Z |
| institution | BVB |
| language | German |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-006952444 |
| oclc_num | 64527236 |
| open_access_boolean | |
| owner | DE-29T DE-703 DE-83 DE-11 DE-188 |
| owner_facet | DE-29T DE-703 DE-83 DE-11 DE-188 |
| physical | XI, 162 S. Ill., graph. Darst. |
| publishDate | 1995 |
| publishDateSearch | 1995 |
| publishDateSort | 1995 |
| record_format | marc |
| spelling | Lange, Rupert Verfasser aut Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules vorgelegt von Rupert Lange 1995 XI, 162 S. Ill., graph. Darst. txt rdacontent n rdamedia nc rdacarrier Heidelberg, Univ., Diss., 1995 Mehrschichten-Perzeptron (DE-588)4354626-2 gnd rswk-swf Backpropagation-Algorithmus (DE-588)4354627-4 gnd rswk-swf Maschinelles Lernen (DE-588)4193754-5 gnd rswk-swf (DE-588)4113937-9 Hochschulschrift gnd-content Mehrschichten-Perzeptron (DE-588)4354626-2 s Maschinelles Lernen (DE-588)4193754-5 s Backpropagation-Algorithmus (DE-588)4354627-4 s DE-604 DNB Datenaustausch application/pdf http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=006952444&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA Inhaltsverzeichnis |
| spellingShingle | Lange, Rupert Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules Mehrschichten-Perzeptron (DE-588)4354626-2 gnd Backpropagation-Algorithmus (DE-588)4354627-4 gnd Maschinelles Lernen (DE-588)4193754-5 gnd |
| subject_GND | (DE-588)4354626-2 (DE-588)4354627-4 (DE-588)4193754-5 (DE-588)4113937-9 |
| title | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules |
| title_auth | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules |
| title_exact_search | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules |
| title_full | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules vorgelegt von Rupert Lange |
| title_fullStr | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules vorgelegt von Rupert Lange |
| title_full_unstemmed | Perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules vorgelegt von Rupert Lange |
| title_short | Perfect learning in neural networks |
| title_sort | perfect learning in neural networks on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object oriented simulation framework for scalable information processing modules |
| title_sub | on sample complexity and scaling issues of gradient descent trained multilayer perceptrons and a novel object-oriented simulation framework for scalable information processing modules |
| topic | Mehrschichten-Perzeptron (DE-588)4354626-2 gnd Backpropagation-Algorithmus (DE-588)4354627-4 gnd Maschinelles Lernen (DE-588)4193754-5 gnd |
| topic_facet | Mehrschichten-Perzeptron Backpropagation-Algorithmus Maschinelles Lernen Hochschulschrift |
| url | http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=006952444&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA |
| work_keys_str_mv | AT langerupert perfectlearninginneuralnetworksonsamplecomplexityandscalingissuesofgradientdescenttrainedmultilayerperceptronsandanovelobjectorientedsimulationframeworkforscalableinformationprocessingmodules |