Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning.

Slides:



Advertisements
Similar presentations
Reinforcement Learning
Advertisements

Markov Decision Process
Reinforcement Learning
Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning.
Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary.
Eick: Reinforcement Learning. Topic 18: Reinforcement Learning 1. Introduction 2. Bellman Update 3. Temporal Difference Learning 4. Discussion of Project1.
Ai in game programming it university of copenhagen Reinforcement Learning [Outro] Marco Loog.
Class Project Due at end of finals week Essentially anything you want, so long as it’s AI related and I approve Any programming language you want In pairs.
Reinforcement Learning
Reinforcement learning (Chapter 21)
1 Reinforcement Learning Introduction & Passive Learning Alan Fern * Based in part on slides by Daniel Weld.
Markov Decision Processes
Planning under Uncertainty
Reinforcement learning
Reinforcement Learning
CS 182/CogSci110/Ling109 Spring 2008 Reinforcement Learning: Algorithms 4/1/2008 Srini Narayanan – ICSI and UC Berkeley.
Cooperative Q-Learning Lars Blackmore and Steve Block Expertness Based Cooperative Q-learning Ahmadabadi, M.N.; Asadpour, M IEEE Transactions on Systems,
Reinforcement Learning Introduction Presented by Alp Sardağ.
1 Kunstmatige Intelligentie / RuG KI Reinforcement Learning Johan Everts.
Ai in game programming it university of copenhagen Reinforcement Learning [Intro] Marco Loog.
CS 188: Artificial Intelligence Fall 2009 Lecture 10: MDPs 9/29/2009 Dan Klein – UC Berkeley Many slides over the course adapted from either Stuart Russell.
Reinforcement Learning Game playing: So far, we have told the agent the value of a given board position. How can agent learn which positions are important?
Making Decisions CSE 592 Winter 2003 Henry Kautz.
Kunstmatige Intelligentie / RuG KI Reinforcement Learning Sander van Dijk.
CS Reinforcement Learning1 Reinforcement Learning Variation on Supervised Learning Exact target outputs are not given Some variation of reward is.
Utility Theory & MDPs Tamara Berg CS Artificial Intelligence Many slides throughout the course adapted from Svetlana Lazebnik, Dan Klein, Stuart.
MAKING COMPLEX DEClSlONS
Reinforcement Learning
Reinforcement Learning
1 ECE-517 Reinforcement Learning in Artificial Intelligence Lecture 7: Finite Horizon MDPs, Dynamic Programming Dr. Itamar Arel College of Engineering.
Learning Theory Reza Shadmehr & Jörn Diedrichsen Reinforcement Learning 1: Generalized policy iteration.
Reinforcement Learning 主講人:虞台文 Content Introduction Main Elements Markov Decision Process (MDP) Value Functions.
© D. Weld and D. Fox 1 Reinforcement Learning CSE 473.
Cooperative Q-Learning Lars Blackmore and Steve Block Multi-Agent Reinforcement Learning: Independent vs. Cooperative Agents Tan, M Proceedings of the.
CHAPTER 16: Reinforcement Learning. Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1) 2 Introduction Game-playing:
1 Introduction to Reinforcement Learning Freek Stulp.
Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning.
MDPs (cont) & Reinforcement Learning
Reinforcement learning (Chapter 21)
Reinforcement Learning AI – Week 22 Sub-symbolic AI Two: An Introduction to Reinforcement Learning Lee McCluskey, room 3/10
Markov Decision Process (MDP)
Reinforcement Learning: Learning algorithms Yishay Mansour Tel-Aviv University.
QUIZ!!  T/F: Optimal policies can be defined from an optimal Value function. TRUE  T/F: “Pick the MEU action first, then follow optimal policy” is optimal.
Possible actions: up, down, right, left Rewards: – 0.04 if non-terminal state Environment is observable (i.e., agent knows where it is) MDP = “Markov Decision.
Reinforcement Learning Guest Lecturer: Chengxiang Zhai Machine Learning December 6, 2001.
Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary.
Reinforcement Learning  Basic idea:  Receive feedback in the form of rewards  Agent’s utility is defined by the reward function  Must learn to act.
1 Passive Reinforcement Learning Ruti Glick Bar-Ilan university.
Artificial Intelligence Ch 21: Reinforcement Learning
Making complex decisions
Reinforcement learning (Chapter 21)
Reinforcement Learning (1)
Reinforcement learning (Chapter 21)
Teaching Style COSC 6368 Teaching Style COSC 6368
Markov Decision Processes
Reinforcement Learning
Reinforcement Learning
Markov Decision Processes
Markov Decision Processes
Reinforcement Learning
Instructors: Fei Fang (This Lecture) and Dave Touretzky
Framework: Agent in State Space
CS 188: Artificial Intelligence Spring 2006
Introduction to Reinforcement Learning and Q-Learning
CS 188: Artificial Intelligence Spring 2006
Markov Decision Processes
Markov Decision Processes
Reinforcement Learning
Presentation transcript:

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Introduction Supervised Learning: Example Class Reinforcement Learning: Situation Reward …

Eick: Reinforcement Learning. Examples Playing chess: Reward comes at end of game Ping-pong: Reward on each point scored Animals: Hunger and pain - negative reward food intake – positive reward

Eick: Reinforcement Learning. Framework: Agent in State Space 123 R=+5 6 R=  9 9 R=  R=+4 5 R= e e s s s nw x/0.7 w n sw x/0.3 n s s Problem: What actions should an agent choose to maximize its rewards ? ne Example: XYZ-World Remark: no terminal states

123 R=+5 6 R=  9 9 R=  R=+4 5 R= e e s s s nw x/0.7 w n sw x/0.3 n s s ne XYZ-World: Discussion Problem 12 (3.3, 0.5) (3.2, -0.5) (0.6, -0.2) Bellman TD P P : Explanation of discrepancies TD for P/Bellman: Most significant discrepancies in states 3 and 8; minor in state 10 P chooses worst successor of 8; should apply operator x instead P should apply w in state 6, but only does it only in 2/3 of the cases; which affects the utility of state 3 The low utility value of state 8 in TD seems to lower the utility value of state 10  only a minor discrepancy I tried hard but: any better explanations?

R= R=  R=  R= R= e e s s s nw x/0.7 w n sw x/0.3 n s s ne XYZ-World: Discussion Problem 12 Bellman Update  =0.2 Discussion on using Bellman Update for Problem 12: No convergence for  =1.0; utility values seem to run away! State 3 has utility 0.58 although it gives a reward of +5 due to the immediate penalty that follows; we were able to detect that. Did anybody run the algorithm for other  e.g. 0.4 or 0.6 values; if yes, did it converge to the same values? Speed of convergence seems to depend on the value of .

123 R=+5 6 R=  9 9 R=  R=+4 5 R= e e s s s nw x/0.7 w n sw x/0.3 n s s ne XYZ-World: Discussion Problem 12 (0.57, -0.65) (-0.50, 0.47) (-0.18, -0.12) TD TD inverse R P : Other observations: The Bellman update did not converge for  =1 The Bellman update converged very fast for  =0.2 Did anybody try other values for  (e.g. 0.6)? The Bellman update suggest a utility value for 3.6 for state 5; what does this tell us about the optimal policy? E.g. is optimal? TD reversed utility values quite neatly when reward were inversed; x become –x+  with  [-0.08,0.08]. (2.98, -2.99)

Eick: Reinforcement Learning. XYZ-World --- Other Considerations R(s) might be known in advance or has to be learnt. R(s) might be probabilistic or not R(s) might change over time --- agent has to adapt. Results of actions might be known in advance or have to be learnt; results of actions can be fixed, or may change over time.

Eick: Reinforcement Learning. Example: The ABC-World 123 R=+5 6 R=  9 9 R=  10 R=+9 8 R=  3 5 R=  4 4 R=  1 7 e e s s s nw x/0.9 w n ne x/0.1 n s n Problem: What actions should an agent choose to maximize its rewards ? ne Remark: no terminal states sw w To be used in Assignment3:

Eick: Reinforcement Learning. Basic Notations and Preview T(s,a,s’) denotes the probability of reaching s’ when using action a in state s; it describes the transition model A policy  specifies what action to take for every possible state s  S R(s) denotes the reward an agent receives in state s Utility-based agents learn an utility function of states uses it to select actions to maximize the expected outcome utility. Q-learning, on the other hand, learns the expected utility of taking a particular action a in a particular state s (Q-value of the pair (s,a) Finally, reflex agents learn a policy that maps directly from states to actions

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Passive Learning We assume the policy Π is fixed. In state s we always execute action Π(s) Rewards are given.

Eick: Reinforcement Learning. Figure 21.1a Terminal State The Agent follows arrows with probability 0.8 and moves right or left of an arrow with probability 0.1; agents are reflected off walls and transferred back to the original state, if they move towards a wall. All non-terminal states have reward  0.04; The two terminal states have rewards +1 and  1.

Eick: Reinforcement Learning. Typical Trials (1,1)  (1,2)  (1,3)  (1,2)  (1,3) …  (4,3) +1 Goal: Use rewards to learn the expected utility U Π (s)

Eick: Reinforcement Learning. Expected Utility U Π (s) = E [ Σ t=0 γ R(s t ) | Π, S 0 = s ] Expected sum of rewards when the policy is followed.

Eick: Reinforcement Learning. Example (1,1)  (1,2)  (1,3)  (2,3)  (3,3)  (4,3) +1 Total reward: (-0.04 x 5) + 1 = 0.80

Eick: Reinforcement Learning. Direct Utility Estimation Convert the problem to a supervised learning problem: (1,1)  U = 0.72 (2,1)  U = 0.68 … Learn to map states to utilities. Problem: utilities are not independent of each other!

Eick: Reinforcement Learning. Bellman Equation Utility values obey the following equations: U  (s) = R(s) + γ  max a Σ s’ T ( s,a,s’ )  U  (s’) Can be solved using dynamic programming. Assumes knowledge of transition model T and reward R; the result is policy independent! Assume γ =1, for this lecture! Incorrect formula replaced on March 10, 2006

Eick: Reinforcement Learning. Example U(1,3) = 0.84 U(2,3) = 0.92 We hope to see that: U(1,3) = U(2,3) The value is Current value is a bit low and we must increase it. (1,3)(2,3)

Eick: Reinforcement Learning. Bellman Update (Section 17.2 of textbook) If we apply the Bellman update indefinitely often, we obtain the utility values that are the solution for the Bellman equation!! U i+1 (s) = R(s) + γ max a ( Σ s’ (T(s, a,s’)  U i (s’))) Some Equations for the XYZ World: U i+1 (1) = 0+ γ*U i (2) U i+1 (5) = 3+ γ *max(U i (7),U i (8)) U i+1 (8) =  + γ *max(U i (6),0.3*U i (7) + 0.7*U i (9) ) Bellman Update:

Eick: Reinforcement Learning. Updating Estimations Based on Observations: New_Estimation = Old_Estimation*(1-  ) + Observed_Value*  New_Estimation= Old_Estimation + Observed_Difference*  Example: Measure the utility of a state s with current value being 2 and observed values are 3 and 3 and the learning rate is 0.2: Initial Utility Value:2 Utility Value after observing 3: 2x x0.2=2.2 Utility Value after observing 3,3: 2.2x0.8 +3x0.2= 2.36

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Temporal Difference Learning Idea: Use observed transitions to adjust values in observed states so that the comply with the constraint equation, using the following update rule: U Π (s)  U Π (s) + α [ R(s) + γ U Π (s’) - U Π (s) ] α is the learning rate; γ discount rate Temporal difference equation. No model assumption --- T and R have not to be known.

Eick: Reinforcement Learning. Fig. 21.5a

Eick: Reinforcement Learning. Fig. 21.5b

Eick: Reinforcement Learning. TD-Q-Learning Goal: Measure the utility of using action a in state s, denoted by Q(a,s); the following update formula is used every time an agent reaches state s’ from s using actions a: Q(a,s)  Q(a,s) + α [ R(s) + γ *max a’ Q(a’,s’)  Q(a,s) ] α is the learning rate;  is the discount factor Variation of TD-Learning Not necessary to know transition model T!

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Active Reinforcement Learning Now we must decide what actions to take. Optimal policy: Choose action with highest utility value. Is that the right thing to do?

Eick: Reinforcement Learning. Active Reinforcement Learning No! Sometimes we may get stuck in suboptimal solutions. Exploration vs Exploitation Tradeoff Why is this important? The learned model is not the same as the true environment.

Eick: Reinforcement Learning. Explore vs Exploit Exploitation: Maximize its reward vs Exploration: Maximize long-term well being.

Eick: Reinforcement Learning. Simple Solution to the Exploitation/Exploration Problem Choose a random action once in k times Otherwise, choose the action with the highest expected utility (k-1 out of k times)

Eick: Reinforcement Learning. Another Solution --- Combining Exploration and Exploitation U + (s)  R(s) + γ  max a f(u,n) u=  s’ (T(s,a,s’)*U + (s’)); n=N(a,s) U + (s) : optimistic estimate of utility N(a,s): number of times action a has been tried. f(u,n): exploration function (idea: returns the value u, if n is large, and values larger than u as n decreases) Example: f(u,n):= if n>n avg then u else max(n/n avg *u, u avg ) n avg being the average number of operator applications. Idea f: Utility of states/actions that have not been explored much is increased artificially.

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Applications Game Playing Checker playing program by Arthur Samuel (IBM) Update rules: change weights by difference between current states and backed-up value generating full look-ahead tree

Eick: Reinforcement Learning. Reinforcement Learning Introduction Passive Reinforcement Learning Temporal Difference Learning Active Reinforcement Learning Applications Summary

Eick: Reinforcement Learning. Summary Goal is to learn utility values of states and an optimal mapping from states to actions. Direct Utility Estimation ignores dependencies among states  we must follow Bellman Equations. Temporal difference updates values to match those of successor states. Active reinforcement learning learns the optimal mapping from states to actions.