MoL-2018-33: Proroković, Krsto (2018) Learning to Decide a Formal Language: A Recurrent Neural Network Approach. [Report]
Preview |
Text
MoL-2018-33.text.pdf Download (1MB) | Preview |
Abstract
We use recurrent neural networks (RNNs) for deciding locally k-testable languages. We show that, when used for deciding languages, RNNs fail to generalise to unseen examples. However, using attention greatly improves the generalisation. We then implement a differentiable version of the scanner used for deciding locally k-testable languages. We show that RNNs are able to store the k-factors in its memory but not arrange then as a look-up table which is necessary for deciding languages specified by multiple k-factors.
| Item Type: | Report |
|---|---|
| Report Nr: | MoL-2018-33 |
| Series Name: | Master of Logic Thesis (MoL) Series |
| Year: | 2018 |
| Subjects: | Computation Logic |
| Depositing User: | Dr Marco Vervoort |
| Date Deposited: | 22 Dec 2018 11:50 |
| Last Modified: | 22 Dec 2018 11:50 |
| URI: | https://eprints.illc.uva.nl/id/eprint/1651 |
Actions (login required)
 |
View Item |
