GRPresentation

View Github Repository Open presentation in a new window

netogallo

See all presentation from netogallo

GRPresentation

0 0

GRPresentation

Presentation for my guided research.

On Github netogallo / GRPresentation

Training Echo State Networks with short segments of Motion Capture Data

Ernesto Rodriguez

Jacobs University Bremen

Supervisor: Dr. Herbert Jaeger

Learning Human Movements

  • ESNs have been successful in learning human motion
  • There exist other methods such as similarity search
  • Human motion is a high degree non-linear signal
  • Can a computer learn motion as efficiently as a human without any context?

Kaiser

\begin{align} \dot{x} & = \sigma(y-x) \ \dot{y} & = \rho x - y - xz \ \dot{z} & = -\beta z + xy \end{align}
\begin{align} \dot{x} & = \sigma(y-x) \\ \dot{y} & = \rho x - y - xz \\ \dot{z} & = -\beta z + xy \end{align}