Markov Decision Process (MDP)

Youtube search... ...Google search

• Markov Model (Chain, Discrete Time, Continuous Time, Hidden)
• Reinforcement Learning (RL)
  • Monte Carlo (MC) Method - Model Free Reinforcement Learning
  • Q Learning
  • State-Action-Reward-State-Action (SARSA)
  • Deep Reinforcement Learning (DRL) DeepRL
    • IMPALA (Importance Weighted Actor-Learner Architecture)
  • Distributed Deep Reinforcement Learning (DDRL)
  • Deep Q Network (DQN)
  • Evolutionary Computation / Genetic Algorithms
  • Asynchronous Advantage Actor Critic (A3C)
  • Hierarchical Reinforcement Learning (HRL)
    • HIerarchical Reinforcement learning with Off-policy correction(HIRO)
  • MERLIN

Solutions:

• Dynamic Programming
• Monte Carlo
• Difference Learning

Used where outcomes are partly random and partly under the control of a decision maker. MDP is a discrete time stochastic control process. At each time step, the process is in some state s, and the decision maker may choose any action a that is available in state s. The process responds at the next time step by randomly moving into a new state s', and giving the decision maker a corresponding reward R_{a}(s,s')} R_a(s,s'). The probability that the process moves into its new state s' is influenced by the chosen action. Helping the convergence of certain algorithms a discount rate (factor) makes an infinite sum finite.