#Solution to Cartpole balancing problem with the help of reinforcement learning and Deep Neural Networks.
I highly recommend going through following blogs and videos if you are new to RL
- Thomos Simonini: Reinforcement Learning with Q Learning: https://lnkd.in/eh46x3V
- Cartpole balancing with q learning: https://medium.com/@tuzzer/cart-pole-balancing-with-q-learning-b54c6068d947
- An overview of reinforcement learning teaching machines to play games: https://pythonmachinelearning.pro/an-overview-of-reinforcement-learning-teaching-machines-to-play-games/
- OpenAI cartpole using Neural Networks by Sentdex : https://pythonprogramming.net/openai-cartpole-neural-network-example-machine-learning-tutorial/
- Reinforcement Learning by Jacob Schrum : https://lnkd.in/eRS2vBY
- Reinforcement Learning by David Silver : https://lnkd.in/ebtdXkd
The code is well explaind by blog no 2 and 3 and it works very well using Q learning algorithm by stabilizing the pole for 200 timsteps I have also tried to solve the problem using Sarsa, it is also working fine but we only have to set the discount rate = 0.98.
A Deep Neural Network can also come up with the solution to this problem, go through Sentdex's website (no. 4) for more info.