| Presentation | 2014-11-17 Reconstructive Neural Net based on Context Free Grammar Takuo HAMAGUCHI, Masahi SHIMBO, Yuji MATSUMOTO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | There are two aspects about Neural Language Model, one aspect, 'Encode', obtains a vector from a sequence and the other aspect, 'Decode', obtains a sequence from a vector. Recurrent Neural Network (Recurrent NN) is a kind of 'Decode' model which repeats prediction and transition from vector-represented state through time. However, a tree structure based model such as Context Free Grammar is more natural for natural languate tha a time-series structure based model. This paper proposes new tree structure 'Decode' Model called Reconstrucitve Neural Network (Reconstructive NN). We conduct the following four experiemtns to show this model's ability: comparision of Recurrent NN and Reconstructive NN, relationship between the accuracy and the Distribution of word length included training data, relationship between the vector dimension and the words length, and the ability of Reconstructive NN comes from natural language. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Deep Learning / Distributed Representation / Recurrent Neural Network / Recursive Nerural Network |
| Paper # | IBISML2014-48 |
| Date of Issue |
| Conference Information | |
| Committee | IBISML |
|---|---|
| Conference Date | 2014/11/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Information-Based Induction Sciences and Machine Learning (IBISML) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Reconstructive Neural Net based on Context Free Grammar |
| Sub Title (in English) | |
| Keyword(1) | Deep Learning |
| Keyword(2) | Distributed Representation |
| Keyword(3) | Recurrent Neural Network |
| Keyword(4) | Recursive Nerural Network |
| 1st Author's Name | Takuo HAMAGUCHI |
| 1st Author's Affiliation | Graduate School of Information Science Nara Institute of Science and Technology() |
| 2nd Author's Name | Masahi SHIMBO |
| 2nd Author's Affiliation | Graduate School of Information Science Nara Institute of Science and Technology |
| 3rd Author's Name | Yuji MATSUMOTO |
| 3rd Author's Affiliation | Graduate School of Information Science Nara Institute of Science and Technology |
| Date | 2014-11-17 |
| Paper # | IBISML2014-48 |
| Volume (vol) | vol.114 |
| Number (no) | 306 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |