The method for handling a transient electromagnetic problem with arbitrary time dependence of a medium parameter is proposed. The method is based on the e¨olutionary approach, which reduces the problem to a Volterra integral equation. A parameter's arbitrary time ¨ariation is approximated by a stepp
Evolutionary reinforcement learning system with time-varying parameters
β Scribed by Kosuke Umesako; Masanao Obayashi; Kunikazu Kobayashi
- Publisher
- John Wiley and Sons
- Year
- 2006
- Tongue
- English
- Weight
- 839 KB
- Volume
- 156
- Category
- Article
- ISSN
- 0424-7760
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
## Abstract Wiener systems which consist of a dynamic linear block followed by a static nonlinear element have been used in numerous applications. In many cases, the system parameters are affected by changes in the environmental conditions. This paper describes a new approach to the onβline identif
A new off-line identification algorithm which combines in a nonlinear fashion the estimates yielded by two Kalman filters running forward and backward in time, respectively, is capable of estimating system parameters varying in a discontinuous way.
The well-known scaled small gain condition guarantees stability for a linear time invariant system subject to bounded complex nonlinear and/or time-varying perturbations. A polynomial time computable condition is derived that can be substantially less conservative for gain scheduled and other multiv
## Abstract According to Hebb's __cell assembly theory__, the brain has the capability of function localization. On the other hand, it is suggested that in the brain there are three different learning paradigms: supervised, unsupervised, and reinforcement learning, which are related deeply to the t