Few-shot text classification with distributional signatures
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, May, 2020
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/1721.1/130200 |
| _version_ | 1826196412626370560 |
|---|---|
| author | Wu, Menghua,M. Eng.Massachusetts Institute of Technology. |
| author2 | Regina Barzilay. |
| author_facet | Regina Barzilay. Wu, Menghua,M. Eng.Massachusetts Institute of Technology. |
| author_sort | Wu, Menghua,M. Eng.Massachusetts Institute of Technology. |
| collection | MIT |
| description | Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, May, 2020 |
| first_indexed | 2024-09-23T10:26:27Z |
| format | Thesis |
| id | mit-1721.1/130200 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T10:26:27Z |
| publishDate | 2021 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1302002021-03-23T03:28:12Z Few-shot text classification with distributional signatures Wu, Menghua,M. Eng.Massachusetts Institute of Technology. Regina Barzilay. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Electrical Engineering and Computer Science. Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, May, 2020 Date of graduation confirmed by MIT Registrar Office. "May 2020." Cataloged from student-submitted PDF of thesis. Includes bibliographical references (pages 18-21). We explore meta-learning for few-shot text classification. Meta-learning has shown strong performance in computer vision, where low-level patterns are transferable across learning tasks. However, directly applying this approach to text is challenging-lexical features highly informative for one task may be insignificant for another. Thus, rather than learning solely from words, our model also leverages their distributional signatures, which encode pertinent word occurrence patterns. Our model is trained within a meta-learning framework to map these signatures into attention scores, which are then used to weight the lexical representations of words. We demonstrate that our model consistently outperforms prototypical networks learned on lexical knowledge (Snell et al., 2017) in both few-shot text classification and relation classification by a significant margin across six benchmark datasets (20.0% on average in 1-shot classification). by Menghua Wu. M. Eng. M.Eng. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science 2021-03-22T17:17:14Z 2021-03-22T17:17:14Z 2020 Thesis https://hdl.handle.net/1721.1/130200 1241198010 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 34 pages application/pdf Massachusetts Institute of Technology |
| spellingShingle | Electrical Engineering and Computer Science. Wu, Menghua,M. Eng.Massachusetts Institute of Technology. Few-shot text classification with distributional signatures |
| title | Few-shot text classification with distributional signatures |
| title_full | Few-shot text classification with distributional signatures |
| title_fullStr | Few-shot text classification with distributional signatures |
| title_full_unstemmed | Few-shot text classification with distributional signatures |
| title_short | Few-shot text classification with distributional signatures |
| title_sort | few shot text classification with distributional signatures |
| topic | Electrical Engineering and Computer Science. |
| url | https://hdl.handle.net/1721.1/130200 |
| work_keys_str_mv | AT wumenghuamengmassachusettsinstituteoftechnology fewshottextclassificationwithdistributionalsignatures |