Lecture 2: Making Sequences of Good Decisions Given a Model of the - PowerPoint PPT Presentation

Lecture 2: Making Sequences of Good Decisions Given a Model of the World Emma Brunskill CS234 Reinforcement Learning Winter 2020 Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 1 / 62

Refresh Your Knowledge 1. Piazza Poll In a Markov decision process, a large discount factor γ means that short term rewards are much more influential than long term rewards. [Enter your answer in piazza] True False Don’t know False. A large γ implies we weigh delayed / long term rewards more. γ = 0 only values immediate rewards Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 2 / 62

Today’s Plan Last Time: Introduction Components of an agent: model, value, policy This Time: Making good decisions given a Markov decision process Next Time: Policy evaluation when don’t have a model of how the world works Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 3 / 62

Models, Policies, Values Model : Mathematical models of dynamics and reward Policy : Function mapping agent’s states to actions Value function : future rewards from being in a state and/or action when following a particular policy Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 4 / 62

Today: Given a model of the world Markov Processes Markov Reward Processes (MRPs) Markov Decision Processes (MDPs) Evaluation and Control in MDPs Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 5 / 62

Full Observability: Markov Decision Process (MDP) MDPs can model a huge number of interesting problems and settings Bandits: single state MDP Optimal control mostly about continuous-state MDPs Partially observable MDPs = MDP where state is history Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 6 / 62

Recall: Markov Property Information state: sufficient statistic of history State s t is Markov if and only if: p ( s t +1 | s t , a t ) = p ( s t +1 | h t , a t ) Future is independent of past given present Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 7 / 62

Markov Process or Markov Chain Memoryless random process Sequence of random states with Markov property Definition of Markov Process S is a (finite) set of states ( s ∈ S ) P is dynamics/transition model that specifices p ( s t +1 = s ′ | s t = s ) Note: no rewards, no actions If finite number ( N ) of states, can express P as a matrix   P ( s 1 | s 1 ) P ( s 2 | s 1 ) · · · P ( s N | s 1 ) P ( s 1 | s 2 ) P ( s 2 | s 2 ) · · · P ( s N | s 2 )     P = . . . ...  . . .  . . .   P ( s 1 | s N ) P ( s 2 | s N ) · · · P ( s N | s N ) Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 8 / 62

Example: Mars Rover Markov Chain Transition Matrix, P ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.2 0.2 0.6 0.2 0.2 0.2   0 . 6 0 . 4 0 0 0 0 0 0 . 4 0 . 2 0 . 4 0 0 0 0     0 0 . 4 0 . 2 0 . 4 0 0 0     P = 0 0 0 . 4 0 . 2 0 . 4 0 0     0 0 0 0 . 4 0 . 2 0 . 4 0     0 0 0 0 0 . 4 0 . 2 0 . 4   0 0 0 0 0 0 . 4 0 . 6 Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 9 / 62

Example: Mars Rover Markov Chain Episodes ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.2 0.2 0.2 0.2 0.2 0.6 Example: Sample episodes starting from S4 s 4 , s 5 , s 6 , s 7 , s 7 , s 7 , . . . s 4 , s 4 , s 5 , s 4 , s 5 , s 6 , . . . s 4 , s 3 , s 2 , s 1 , . . . Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 10 / 62

Markov Reward Process (MRP) Markov Reward Process is a Markov Chain + rewards Definition of Markov Reward Process (MRP) S is a (finite) set of states ( s ∈ S ) P is dynamics/transition model that specifices P ( s t +1 = s ′ | s t = s ) R is a reward function R ( s t = s ) = E [ r t | s t = s ] Discount factor γ ∈ [0 , 1] Note: no actions If finite number ( N ) of states, can express R as a vector Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 11 / 62

Example: Mars Rover MRP ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.2 0.2 0.6 0.2 0.2 0.2 Reward: +1 in s 1 , +10 in s 7 , 0 in all other states Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 12 / 62

Return & Value Function Definition of Horizon Number of time steps in each episode Can be infinite Otherwise called finite Markov reward process Definition of Return, G t (for a MRP) Discounted sum of rewards from time step t to horizon G t = r t + γ r t +1 + γ 2 r t +2 + γ 3 r t +3 + · · · Definition of State Value Function, V ( s ) (for a MRP) Expected return from starting in state s V ( s ) = E [ G t | s t = s ] = E [ r t + γ r t +1 + γ 2 r t +2 + γ 3 r t +3 + · · · | s t = s ] Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 13 / 62

Discount Factor Mathematically convenient (avoid infinite returns and values) Humans often act as if there’s a discount factor < 1 γ = 0: Only care about immediate reward γ = 1: Future reward is as beneficial as immediate reward If episode lengths are always finite, can use γ = 1 Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 14 / 62

Example: Mars Rover MRP ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.2 0.2 0.2 0.2 0.6 0.2 Reward: +1 in s 1 , +10 in s 7 , 0 in all other states Sample returns for sample 4-step episodes, γ = 1 / 2 s 4 , s 5 , s 6 , s 7 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 10 = 1 . 25 Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 15 / 62

Example: Mars Rover MRP ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.2 0.2 0.6 0.6 0.2 0.2 0.2 Reward: +1 in s 1 , +10 in s 7 , 0 in all other states Sample returns for sample 4-step episodes, γ = 1 / 2 s 4 , s 5 , s 6 , s 7 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 10 = 1 . 25 s 4 , s 4 , s 5 , s 4 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 0 = 0 s 4 , s 3 , s 2 , s 1 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 1 = 0 . 125 Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 16 / 62

Example: Mars Rover MRP ! " ! # ! $ ! % ! & ! ' ! ( 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.2 0.6 0.6 0.2 0.2 0.2 0.2 Reward: +1 in s 1 , +10 in s 7 , 0 in all other states Value function: expected return from starting in state s V ( s ) = E [ G t | s t = s ] = E [ r t + γ r t +1 + γ 2 r t +2 + γ 3 r t +3 + · · · | s t = s ] Sample returns for sample 4-step episodes, γ = 1 / 2 s 4 , s 5 , s 6 , s 7 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 10 = 1 . 25 s 4 , s 4 , s 5 , s 4 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 0 = 0 s 4 , s 3 , s 2 , s 1 : 0 + 1 2 × 0 + 1 4 × 0 + 1 8 × 1 = 0 . 125 V = [1.53 0.37 0.13 0.22 0.85 3.59 15.31] Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 17 / 62

Computing the Value of a Markov Reward Process Could estimate by simulation Generate a large number of episodes Average returns Concentration inequalities bound how quickly average concentrates to expected value Requires no assumption of Markov structure Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 18 / 62

Computing the Value of a Markov Reward Process Could estimate by simulation Markov property yields additional structure MRP value function satisfies � P ( s ′ | s ) V ( s ′ ) V ( s ) = R ( s ) + γ ���� s ′ ∈ S Immediate reward � �� � Discounted sum of future rewards Emma Brunskill (CS234 Reinforcement Learning) Lecture 2: Making Sequences of Good Decisions Given a Model of the World Winter 2020 19 / 62