Work Package: WP1
Deliverable: Deliverable 2.1: “Design of a new method of signal processing for robotics or BCI”
Deliverable due date: month 30
This document reports on the progress of work on Deliverable 1.2: “P1 will comprise the modelling of a new method of signal processing and the adaptation and validation of such a model for robotics or BCI. The objective is to integrate this model into commercial software.”
All the planned tasks related to this Deliverable have been accomplished. Obtained results have been submitted to PLOS Computational Biology” peer-reviewed journal in 2015 and is under peer-review. The publication is a scientific part of the Deliverable 2.1 and this “Project Deliverable Report” document focuses on its relation to the NETT proposal.
Neural networks in the brain process signals using spikes. The big advantage of this sparse communication principle is its energy efficiency and the small amount of bandwidth needed between the single computing units (neurons). Using principles of neural coding we redesigned a machine learning method for learning online computations on time varying signals (echo state networks) and developed a spike based version of it.
In such a way we accomplish Deliverable 2.1. The second part of P2 “Software validation, integration into existing technology” is work in progress.
Contributors: Dominik Thalmeier, Bert Kappen (RU)