Abstract a paradigm of modern Machine Learning (ML) which uses rewards and punishments to guide the learning process. One of the central ideas of RL is learning by “direct-online...
The paper presents and evaluates the power of best-first search over AND/OR search spaces in graphical models. The main virtue of the AND/OR representation is its sensitivity to ...
Abstract— The human communication exists in various situations of our daily life. For human – robot communication or robot – robot communication, it is useful to design a com...
We develop a dynamic optimal control model of a fashion designer's challenge of maintaining brand image in the face of short-term pro...t opportunities through expanded sales...
Peter M. Kort, Jonathan P. Caulkins, Richard F. Ha...
This work describes a forward-looking approach for the solution of dynamic (time-changing) problems using evolutionary algorithms. The main idea of the proposed method is to combi...