Skip to content

jvitku/designer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

97 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Set of Algorithms Purposed to Design Hybrid Artificial Neural Network Systems

Author Jaroslav Vitku [vitkujar@fel.cvut.cz]

About

This Designer is used for optimization of topology of in Hybrid Artificial Neural Network Systems, used in the Nengoros simulator.

This project serves as place for first types of algorithms used for designing hybrid systems, that is: designing topologies of Hybrid Artificial Neural Network Systems (HANNS).

The first project - basicea implements basic Evolutionary Algorithm (EA), which optimizes vector of real-valued numbers. These numbers represent specific connection weights in an agent architecture. By modifying these weights, new HANNS topology is created - new agent architecture is designed.

Requirements

Complete list of dependencies can be found in the build.gradle file, some of the main requirements are:

Installation

Installation can be obtained by running the following command

./gradlew installApp

In case of any problems, the best way how to use these nodes is by means of the NengoRos project (see: https://github.com/jvitku/nengoros )

About

holds EAs for designing of hybrid networks in the NengoROS simulator

Resources

Stars

Watchers

Forks

Packages

No packages published