Skip to content

Latest commit

 

History

History
7 lines (4 loc) · 624 Bytes

File metadata and controls

7 lines (4 loc) · 624 Bytes

Visual-Agent

C++ implementation of Randy Beer's VisualAgent brain-body-environment model: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.120.3706&rep=rep1&type=pdf

This agent is controlled by a continuous time recurrent neural network (CTRNN) and is capable of performing categorical perception given visual feedback in a dynamic environment.

You can find the CTRNN implementation, and an implementation of evolutionary algorithms for training the networks here: http://pages.iu.edu/~rdbeer/ (scroll down to the Evolutionary Agents section). Weights for a pre-trained CTRNN can be found in categorize.ns.txt.