A study of recurrent neural networks for grammatical inference
The objective of this thesis is twofold. Firstly, we want to study the potential of recurrent neural networks as inference engines for discovering rules generated by unknown grammars under the formal language hierarchy. Secondly, we wish to explore the (in)capabilities of recurrent networks for lear...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2009
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/19577 |
| Summary: | The objective of this thesis is twofold. Firstly, we want to study the potential of recurrent neural networks as inference engines for discovering rules generated by unknown grammars under the formal language hierarchy. Secondly, we wish to explore the (in)capabilities of recurrent networks for learning grammars, proposing enhanced models to rectify any identified limitations. |
|---|