Abstract
dc:description.abstractCapsule Networks exploit a new, vector-based perceptron model, providing feature instantiation parameters on top of feature existence probabilities. With these vectors, simple scalar operations are elaborated to vector-matrix multiplication and multi-vector weighted reduction. Capsule Networks include convolutional layers which take the initial input and help it become a tensor.A novel data abstraction maps the individual values of this tensor to one-dimensional arrays but is conceptualized as a 2D grids of multi-dimensional elements. Moreover, loss function thresholds and architectural dimensions were arbitrarily set during the introduction of Capsule Networks. While current machine learning libraries provide abstractions for convolutional layers, a TensorFlow optimization requires structural overhead for a full Capsule Network implementation. They lack simple optimizations specifically for Capsule Network data allocation. This thesis presents a scalable GPU optimization for the training and evaluation of Capsule Networks. Furthermore, hyperparameters are adjusted with the help of a multi-objective evolutionary algorithm (MOEA) to minimize the loss function while maximizing accuracy.
Degree
thesis:*- Level thesis:degree_level
- Master's Degree
- Department thesis:degree_department
- Computer Science and Engineering
- Year dc:date.issued
- 2018
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Lopez, Daniel A.
- Advisor dc:contributor.advisor
-
- Harris, Frederick C.
- Committee members dc:contributor.committeemember
-
- Latourrette, Nancy
- Mortensen, Jeff
Subjects
dc:subject × 5Rights
dc:rights- Statement dc:rights
-
- Creative Commons Attribution-ShareAlike 4.0 United States
- Author(s)
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- http://hdl.handle.net/11714/4551