Fault Tolerance of Sigmoidal Feedforward Artificial Neural Networks
| dc.contributor.guide | C.S. Rai and Pravin Chandra | |
| dc.coverage.spatial | ||
| dc.creator.researcher | Amit Prakash Singh | |
| dc.date.accessioned | 2017-12-27T10:10:47Z | |
| dc.date.available | 2017-12-27T10:10:47Z | |
| dc.date.awarded | 01/12/2011 | |
| dc.date.completed | 2011 | |
| dc.date.registered | 2004 | |
| dc.description.abstract | Artificial neural networks (ANN) are known for their parallel structure. Based on the analogy with the biological neural structure, the robustness of the ANN working under faulty conditions has been claimed. That is, it has been claimed that ANNs working under fault conditions degrades gracefully. With the advent of ANN implementations in hardware it has become imperative to study the fault (tolerant) characteristics of ANNs. The following properties may be inferred for ANNs (based on the analogy to the biological neural systems): newline The computation is massively distributed and parallel. newline Integration and storage of experience, which could be previous classification or association of learning data or, self - organization of experience. newline Consideration of new experience in the context of stored experience or, a newlinecontext - addressable structure. newline Ability to make accurate predictions about new situations on the basis of previously (self-) organized experiences or, generalization capability. newline A fault - tolerant architecture, in the sense that loss of a few neurons may be recoverable by adaptation of those remaining (perhaps with additional training) and/or by growth of new neurons or associations between neurons. newline... | |
| dc.description.note | ||
| dc.format.accompanyingmaterial | CD | |
| dc.format.dimensions | ||
| dc.format.extent | ||
| dc.identifier.uri | http://hdl.handle.net/10603/184849 | |
| dc.language | English | |
| dc.publisher.institution | University School of Information and Communication Technology | |
| dc.publisher.place | Delhi | |
| dc.publisher.university | Guru Gobind Singh Indraprastha University | |
| dc.relation | ||
| dc.rights | university | |
| dc.source.university | University | |
| dc.title | Fault Tolerance of Sigmoidal Feedforward Artificial Neural Networks | |
| dc.title.alternative | ||
| dc.type.degree | Ph.D. |
Files
Original bundle
1 - 5 of 19
Loading...
- Name:
- 01 title page certificate acknowledgment.pdf
- Size:
- 1.22 MB
- Format:
- Adobe Portable Document Format
- Description:
- Attached File
Loading...
- Name:
- 04 list of figure.pdf
- Size:
- 2.32 MB
- Format:
- Adobe Portable Document Format
Loading...
- Name:
- 05 list of tables.pdf
- Size:
- 5.18 MB
- Format:
- Adobe Portable Document Format
License bundle
1 - 1 of 1