A Study on Real World Experience Learning and Solving Symbol Grounding using Mutual Transfer of Multimodal Information
| Project/Area Number |
15K12104
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Soft computing
|
|---|
| Research Institution | Ochanomizu University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 潜在状態推定 / ガウス過程 / 時系列データ / 多層パーセプトロン / SharedGPLVM / SharedGPDM / GPLVM / GPDM / 潜在空間 / 動作認識 / 経験学習 / ヒューマノイドロボット |
|---|
| Outline of Final Research Achievements |
In this study, we have proposed an efficient method to obtain the latent state of multi-dimensional time-series data following Gaussian process, using Gaussian Process Latent Variable Model and Gaussian Process Dynamic Model as the base methods. Furthermore, as a method to achieve high efficient and precision for estimating the latent states of the data, we introduced Multi-layer Perceptron to estimate the states instead of using EM algorithm. We applied our proposed method to SharedGPLVM and SharedGPDM, and have confirmed that our proposed method works to estimate the latent states with high efficiency and precision.
|
Report
(4 results)
Research Products
(10 results)
