Contains Solutions of Lab assignments of Reinforcement Learning lab

see wiki for documentation and getting started.

Index:

Lab1:

  Probability And Statistics:
  Markov Chain, Sampling from Distributions,

Lab2:

  Multi Arm Bandits:
  Study of algorithms Like UCB, Thompson Sampling, Epsilon Greedy, Reinforce, Softmax for Multi Arm Bandits Problem with Bernoulli and Gaussian reward distribution.

Lab3:

  DP Methods for RL:
  Policy And Value Iteration for GridWorld

Lab4:

  Model Free RL Algorithms:
  MonteCarlo Control, SARSA, Q-Learning for MountainCar (Continious env), Taxi (discrete env).

Lab5:

  Linear Function Approximation and Policy Gradients:
  ~MonteCarlo Control, SARSA, Q-Learning with function approximation~, DQN and A2C

Mini Project:

  Literature survey, implementation and evaluation of Proximal Policy Optimization for various tasks.

Others:

  Other codes and assignments following various MOOCs.