Proposal of an Exploitation-oriented Learning Method on Multiple Rewards and Penalties Environments and the Design Guideline.
Kazuteru MiyazakiPublished in: J. Comput. (2013)
Keyphrases
- high precision
- similarity measure
- preprocessing
- prior knowledge
- objective function
- unsupervised learning
- edge detection
- computational cost
- learning scheme
- dynamic programming
- segmentation method
- dynamic environments
- knowledge based systems
- cost function
- learning process
- machine learning
- learning algorithm
- neural nets
- multiple tasks
- learning mechanism
- design process
- high accuracy
- support vector machine
- user interface
- case study