Continuous Neural Association for Efficient Robot Learning


Invited talk

Speaker(s) : Jochen Steil

Event : Invited seminar

Place : Frankfurt Institute of Advanced Studies

Date and Time : 01/14/2014, 1:00 pm

Abstract : How to represent movement skills in complex behavioral
architectures ?
This is a persistant research question in cognitive robotics which we tackle
through a neuro-robotics approach based on the principle of continous neural
association. The latter is an original approach to combine ideas from
reservoir computing, dynamical systems and classical associative memories in
a coherent framework.
The goal is to learn to bind together sensori-motor data, parametric
representations like dynamic movement primitives and  low-dimensional
embeddings of task-specifying parameters
like e.g. via points in a multi-level and multi-scale skill memory.
It is further shown that this paramteric skill memory is
highly beneficial for application of modern trajectory based learning
  based on roll-outs and reward-weighted averaging, which we demonstrate in
applications to velocity field learning, inverse kinematics control, and
skill optimization for humanoid robots like the iCub.

Partners : Bielefeld University - CoR-Lab


Jochen Steil


01/14/2014, 1:00 pm


Frankfurt Institute of Advanced Studies


Scientific Community

Recent Publications