Reinforcement-Learning

Implementation of Reinforcement Learning Algorithms, mainly in openAI gym

Algorithms Implemented so far:

• Q-learning with Softmax Action Selection: qlearn.py
• Q-learning with e-greedy Action Selection: qlearn2.py
• SARSA with with Softmax Action Selection: sarsa.py

Evaluations:

• qlearn2.py: https://gym.openai.com/evaluations/eval_S6ldgxysSay2eidRk9eHNA
• sarsa.py: https://gym.openai.com/evaluations/eval_D7pDK88ESHCdiZR7YZeqMQ

Running:

openAI gym, numpy, and python3 are required to run any of the programs. To run enter: python3 filename.py on command line