| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2020: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
|
|---|
| Outline of Final Research Achievements |
We replaced the inner product in dictionary learning with a positive-definite kernel to obtain atoms that have a non-linear relationship with the observations. The atoms are templates used for generating observations, hence corresponds to latent features.
Using dictionary learning, we developed a system that extracts features from images. As a specific example, images in a calcium imaging dataset obtained from animal brains were separated into spatial and temporal components. They were examined by neurophysiologists to correspond to patterns of neural activation. Dictionary learning is much faster than training a deep neural network. The developed method can be used for real-time pattern detection, enabling a feedback-control system connected to an animal brain.
|
