This repository contains an implementation of feedback capacity estimator of an unifilar finite-state-channel using reinforcement learning, specifically policy optimization technique.
The code is compatible with a tensorflow 2.0 environment. If you use a docker, you can pull the following docker image
docker pull tensorflow/tensorflow:latest-gpu-py3
The parameters of the channels are in the file example.json
. Modify this file to set your own values for the parameters.
To run the code:
python ./example.py --exp_name <simulation_name> --config ./configs/example.json
- Ziv Aharoni
- Oron Sabag
- Haim Permuter
This project is licensed under Apache License 2.0 - see the LICENSE.md file for details