Physical symmetry enhanced neural networks
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/1721.1/128294 |
| _version_ | 1811077461912322048 |
|---|---|
| author | Jing, Li,Ph. D.Massachusetts Institute of Technology. |
| author2 | Marin Soljacic. |
| author_facet | Marin Soljacic. Jing, Li,Ph. D.Massachusetts Institute of Technology. |
| author_sort | Jing, Li,Ph. D.Massachusetts Institute of Technology. |
| collection | MIT |
| description | This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. |
| first_indexed | 2024-09-23T10:43:25Z |
| format | Thesis |
| id | mit-1721.1/128294 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T10:43:25Z |
| publishDate | 2020 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1282942020-11-04T03:25:39Z Physical symmetry enhanced neural networks Jing, Li,Ph. D.Massachusetts Institute of Technology. Marin Soljacic. Massachusetts Institute of Technology. Department of Physics. Massachusetts Institute of Technology. Department of Physics Physics. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Thesis: Ph. D., Massachusetts Institute of Technology, Department of Physics, February, 2020 Cataloged from student-submitted PDF version of thesis Includes bibliographical references (pages 91-99). Artificial Intelligence (AI), widely considered "the fourth industrial revolution", has shown its potential to fundamentally change our world. Today's AI technique relies on neural networks. In this thesis, we propose several physical symmetry enhanced neural network models. We first developed unitary recurrent neural networks (RNNs) that solve gradient vanishing and gradient explosion problems. We propose an efficient parametrization method that requires [sigma] (1) complexity per parameter. Our unitary RNN model has shown optimal long-term memory ability. Next, we combine the above model with a gated mechanism. This model outperform popular recurrent neural networks like long short-term memory (LSTMs) and gated recurrent units (GRUs) in many sequential tasks. In the third part, we develop a convolutional neural network architecture that achieves logarithmic scale complexity using symmetry breaking concepts. We demonstrate that our model has superior performance on small image classification tasks. In the last part, we propose a general method to extend convolutional neural networks' inductive bias and embed other types of symmetries. We show that this method improves prediction performance on lens-distorted image by Li Jing. Ph. D. Ph.D. Massachusetts Institute of Technology, Department of Physics 2020-11-03T20:28:36Z 2020-11-03T20:28:36Z 2020 2020 Thesis https://hdl.handle.net/1721.1/128294 1201326165 eng MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided. http://dspace.mit.edu/handle/1721.1/7582 99 pages application/pdf Massachusetts Institute of Technology |
| spellingShingle | Physics. Jing, Li,Ph. D.Massachusetts Institute of Technology. Physical symmetry enhanced neural networks |
| title | Physical symmetry enhanced neural networks |
| title_full | Physical symmetry enhanced neural networks |
| title_fullStr | Physical symmetry enhanced neural networks |
| title_full_unstemmed | Physical symmetry enhanced neural networks |
| title_short | Physical symmetry enhanced neural networks |
| title_sort | physical symmetry enhanced neural networks |
| topic | Physics. |
| url | https://hdl.handle.net/1721.1/128294 |
| work_keys_str_mv | AT jingliphdmassachusettsinstituteoftechnology physicalsymmetryenhancedneuralnetworks |