Multiple Model Kalman Temporal Differences (MM-KTD) framework for value function approximation.
This repository contains the codes for the MM-KTD framework, introduced in the paper:
Multiple Model Kalman Temporal Differences for Reinforcement Learning (https://arxiv.org/pdf/2006.00195.pdf)
Figs directory includes all of the figures in the paper.
gym-foo directory is a fork from https://github.com/openai/gym/blob/master/docs/creating-environments.md.
For the Inverted Pendulum experiment, run Inverted_Pendulum.py in the MM-KTD codes directory.
For the Lunar Lander experiment, run Lunar_lander.py in the MM-KTD codes directory.
For the Mountain Car experiment, run Mountain_Car.py in the MM-KTD codes directory.